Datasets:

ppxscal

/

academic_embeddings_cosimrank_test

like 0

ROD-Revenue: Seeking Strategies Analysis and Revenue Prediction in Ride-on-Demand Service Using Multi-Source Urban Data Recent years have witnessed the rapidly-growing business of ride-on-demand (RoD) services such as Uber, Lyft and Didi. Unlike taxi services, these emerging transportation services use dynamic pricing to manipulate the supply and demand, and to improve service responsiveness and quality. Despite this, on the drivers' side, dynamic pricing creates a new problem: how to seek for passengers in order to earn more under the new pricing scheme. Seeking strategies have been studied extensively in traditional taxi service, but in RoD service such studies are still rare and require the consideration of more factors such as dynamic prices, the status of other transportation services, etc. In this paper, we develop ROD-Revenue, aiming to mine the relationship between driver revenue and factors relevant to seeking strategies, and to predict driver revenue given features extracted from multi-source urban data. We extract basic features from multiple datasets, including RoD service, taxi service, POI information, and the availability of public transportation services, and then construct composite features from basic features in a product-form. The desired relationship is learned from a linear regression model with basic features and high-dimensional composite features. The linear model is chosen for its interpretability-to quantitatively explain the desired relationship. Finally, we evaluate our model by predicting drivers' revenue. We hope that ROD-Revenue not only serves as an initial analysis of seeking strategies in RoD service, but also helps increasing drivers' revenue by offering useful guidance.

Weakly Supervised Joint Sentiment-Topic Detection from Text Sentiment analysis or opinion mining aims to use automated tools to detect subjective information such as opinions, attitudes, and feelings expressed in text. This paper proposes a novel probabilistic modeling framework called joint sentiment-topic (JST) model based on latent Dirichlet allocation (LDA), which detects sentiment and topic simultaneously from text. A reparameterized version of the JST model called Reverse-JST, obtained by reversing the sequence of sentiment and topic generation in the modeling process, is also studied. Although JST is equivalent to Reverse-JST without a hierarchical prior, extensive experiments show that when sentiment priors are added, JST performs consistently better than Reverse-JST. Besides, unlike supervised approaches to sentiment classification which often fail to produce satisfactory performance when shifting to other domains, the weakly supervised nature of JST makes it highly portable to other domains. This is verified by the experimental results on data sets from five different domains where the JST model even outperforms existing semi-supervised approaches in some of the data sets despite using no labeled documents. Moreover, the topics and topic sentiment detected by JST are indeed coherent and informative. We hypothesize that the JST model can readily meet the demand of large-scale sentiment analysis from the web in an open-ended fashion.

Twitter in academic events: A study of temporal usage, communication, sentimental and topical patterns in 16 Computer Science conferences •Analysis of Twitter on 16 CS conferences over five years.•Over time, users increase informational use and decrease conversational usage.•LDA allows conference clustering, unveiling shared areas of interest.•Sentiment analysis exposes differences between research communities.•Model of user participation explains the importance of different social features.

A Continuous Representation of Ad Hoc Ridesharing Potential. Interacting with ridesharing systems is a complex spatiotemporal task. Traditional approaches rely on the full disclosure of a client's trip information to perform ride matching. However during poor service conditions of low supply or high demand, this requirement may mean that a client cannot find any ride matching their intentions. To address this within real-world road networks, we extend our map-based opportunistic client user interface concept, i.e., launch pads, from a discrete to a continuous space–time representation of vehicle accessibility to provide a client with a more realistic choice set. To examine this extension under different conditions, we conduct two computational experiments. First, we extend our previous investigation into the effects of varying vehicle flexibility and population size on launch pads and a client's probability of pick-up, describing the increased opportunity. Second, observing launch pads within a real-world road network, we analyze aspects of choice and propose necessary architecture improvements. The communication of ride share potential using launch pads provides a client with a simple yet flexible means of interfacing with on-demand transportation.

Taxi Drivers' Cruising Patterns - Insights from Taxi GPS Traces. In this paper, we seek to identify the different impacts of external and internal information on taxis' cruising behaviors and to find effective methods to enhance taxis' cruising efficiency. Through global positioning system trajectory data collected in Shenzhen, China, we determine the impacts of external factors (land use, traffic conditions, and road grade) and internal factors (previous pick-...

Computational thinking Summary form only given. My vision for the 21st century, Computational Thinking, will be a fundamental skill used by everyone in the world. To reading, writing, and arithmetic, we should add computational thinking to every child's analytical ability. Computational thinking involves solving problems, designing systems, and understanding human behavior by drawing on the concepts fundamental to computer science. Thinking like a computer scientist means more than being able to program a computer. It requires the ability to abstract and thus to think at multiple levels of abstraction. In this talk I will give many examples of computational thinking, argue that it has already influenced other disciplines, and promote the idea that teaching computational thinking can not only inspire future generations to enter the field of computer science but benefit people in all fields.

JPEG Error Analysis and Its Applications to Digital Image Forensics JPEG is one of the most extensively used image formats. Understanding the inherent characteristics of JPEG may play a useful role in digital image forensics. In this paper, we introduce JPEG error analysis to the study of image forensics. The main errors of JPEG include quantization, rounding, and truncation errors. Through theoretically analyzing the effects of these errors on single and double JPEG compression, we have developed three novel schemes for image forensics including identifying whether a bitmap image has previously been JPEG compressed, estimating the quantization steps of a JPEG image, and detecting the quantization table of a JPEG image. Extensive experimental results show that our new methods significantly outperform existing techniques especially for the images of small sizes. We also show that the new method can reliably detect JPEG image blocks which are as small as 8 × 8 pixels and compressed with quality factors as high as 98. This performance is important for analyzing and locating small tampered regions within a composite image.

Highly dynamic Destination-Sequenced Distance-Vector routing (DSDV) for mobile computers An ad-hoc network is the cooperative engagement of a collection of Mobile Hosts without the required intervention of any centralized Access Point. In this paper we present an innovative design for the operation of such ad-hoc networks. The basic idea of the design is to operate each Mobile Host as a specialized router, which periodically advertises its view of the interconnection topology with other Mobile Hosts within the network. This amounts to a new sort of routing protocol. We have investigated modifications to the basic Bellman-Ford routing mechanisms, as specified by RIP [5], to make it suitable for a dynamic and self-starting network mechanism as is required by users wishing to utilize ad hoc networks. Our modifications address some of the previous objections to the use of Bellman-Ford, related to the poor looping properties of such algorithms in the face of broken links and the resulting time dependent nature of the interconnection topology describing the links between the Mobile Hosts. Finally, we describe the ways in which the basic network-layer routing can be modified to provide MAC-layer support for ad-hoc networks.

Efficient Signature Generation by Smart Cards We present a new public-key signature scheme and a corresponding authentication scheme that are based on discrete logarithms in a subgroup of units in Zp where p is a sufficiently large prime, e.g., p = 2512. A key idea is to use for the base of the discrete logarithm an integer a in Zp such that the order of a is a sufficiently large prime q, e.g., q = 2140. In this way we improve the ElGamal signature scheme in the speed of the procedures for the generation and the verification of signatures and also in the bit length of signatures. We present an efficient algorithm that preprocesses the exponentiation of a random residue modulo p.

Stabilizing a linear system by switching control with dwell time The use of networks in control systems to connect controllers and sensors/actuators has become common practice in many applications. This new technology has also posed a theoretical control problem of how to use the limited data rate of the network effectively. We consider a system where its sensor and actuator are connected by a finite data rate channel. A design method to stabilize a continuous-time, linear plant using a switching controller is proposed. In particular, to prevent the actuator from fast switching, or chattering, which can not only increase the necessary data rate but also damage the system, we employ a dwell-time switching scheme. It is shown that a systematic partition of the state-space enables us to reduce the complexity of the design problem

Empirical Modelling of Genetic Algorithms This paper addresses the problem of reliably setting genetic algorithm parameters for consistent labelling problems. Genetic algorithm parameters are notoriously difficult to determine. This paper proposes a robust empirical framework, based on the analysis of factorial experiments. The use of a graeco-latin square permits an initial study of a wide range of parameter settings. This is followed by fully crossed factorial experiments with narrower ranges, which allow detailed analysis by logistic regression. The empirical models derived can be used to determine optimal algorithm parameters and to shed light on interactions between the parameters and their relative importance. Re-fined models are produced, which are shown to be robust under extrapolation to up to triple the problem size.

Beamforming for MISO Interference Channels with QoS and RF Energy Transfer We consider a multiuser multiple-input single-output interference channel where the receivers are characterized by both quality-of-service (QoS) and radio-frequency (RF) energy harvesting (EH) constraints. We consider the power splitting RF-EH technique where each receiver divides the received signal into two parts a) for information decoding and b) for battery charging. The minimum required power that supports both the QoS and the RF-EH constraints is formulated as an optimization problem that incorporates the transmitted power and the beamforming design at each transmitter as well as the power splitting ratio at each receiver. We consider both the cases of fixed beamforming and when the beamforming design is incorporated into the optimization problem. For fixed beamforming we study three standard beamforming schemes, the zero-forcing (ZF), the regularized zero-forcing (RZF) and the maximum ratio transmission (MRT); a hybrid scheme, MRT-ZF, comprised of a linear combination of MRT and ZF beamforming is also examined. The optimal solution for ZF beamforming is derived in closed-form, while optimization algorithms based on second-order cone programming are developed for MRT, RZF and MRT-ZF beamforming to solve the problem. In addition, the joint-optimization of beamforming and power allocation is studied using semidefinite programming (SDP) with the aid of rank relaxation.

An efficient scheduling scheme for mobile charger in on-demand wireless rechargeable sensor networks. Existing studies on wireless sensor networks (WSNs) have revealed that the limited battery capacity of sensor nodes (SNs) hinders their perpetual operation. Recent findings in the domain of wireless energy transfer (WET) have attracted a lot of attention of academia and industry to cater the lack of energy in the WSNs. The main idea of WET is to restore the energy of SNs using one or more wireless mobile chargers (MCs), which leads to a new paradigm of wireless rechargeable sensor networks (WRSNs). The determination of an optimal order of charging the SNs (i.e., charging schedule) in an on-demand WRSN is a well-known NP-hard problem. Moreover, care must be taken while designing the charging schedule of an MC as requesting SNs introduce both spatial and temporal constraints. In this paper, we first present a Linear Programming (LP) formulation for the problem of scheduling an MC and then propose an efficient solution based on gravitational search algorithm (GSA). Our method is presented with a novel agent representation scheme and an efficient fitness function. We perform extensive simulations on the proposed scheme to demonstrate its effectiveness over two state-of-the-art algorithms, namely first come first serve (FCFS) and nearest job next with preemption (NJNP). The simulation results reveal that the proposed scheme outperforms both the existing algorithms in terms of charging latency. The virtue of our scheme is also proved by the well-known statistical test, analysis of variance (ANOVA), followed by post hoc analysis.

Hardware Circuits Design and Performance Evaluation of a Soft Lower Limb Exoskeleton Soft lower limb exoskeletons (LLEs) are wearable devices that have good potential in walking rehabilitation and augmentation. While a few studies focused on the structure design and assistance force optimization of the soft LLEs, rarely work has been conducted on the hardware circuits design. The main purpose of this work is to present a new soft LLE for walking efficiency improvement and introduce its hardware circuits design. A soft LLE for hip flexion assistance and a hardware circuits system with scalability were proposed. To assess the efficacy of the soft LLE, the experimental tests that evaluate the sensor data acquisition, force tracking performance, lower limb muscle activity and metabolic cost were conducted. The time error in the peak assistance force was just 1%. The reduction in the normalized root-mean-square EMG of the rectus femoris was 7.1%. The net metabolic cost in exoskeleton on condition was reduced by 7.8% relative to walking with no exoskeleton. The results show that the designed hardware circuits can be applied to the soft LLE and the soft LLE is able to improve walking efficiency of wearers.

Neural-Network-Based Models for Short-Term Traffic Flow Forecasting Using a Hybrid Exponential Smoothing and Levenberg–Marquardt Algorithm This paper proposes a novel neural network (NN) training method that employs the hybrid exponential smoothing method and the Levenberg–Marquardt (LM) algorithm, which aims to improve the generalization capabilities of previously used methods for training NNs for short-term traffic flow forecasting. The approach uses exponential smoothing to preprocess traffic flow data by removing the lumpiness from collected traffic flow data, before employing a variant of the LM algorithm to train the NN weights of an NN model. This approach aids NN training, as the preprocessed traffic flow data are more smooth and continuous than the original unprocessed traffic flow data. The proposed method was evaluated by forecasting short-term traffic flow conditions on the Mitchell freeway in Western Australia. With regard to the generalization capabilities for short-term traffic flow forecasting, the NN models developed using the proposed approach outperform those that are developed based on the alternative tested algorithms, which are particularly designed either for short-term traffic flow forecasting or for enhancing generalization capabilities of NNs.

Estimation of prediction error by using K-fold cross-validation Estimation of prediction accuracy is important when our aim is prediction. The training error is an easy estimate of prediction error, but it has a downward bias. On the other hand, K-fold cross-validation has an upward bias. The upward bias may be negligible in leave-one-out cross-validation, but it sometimes cannot be neglected in 5-fold or 10-fold cross-validation, which are favored from a computational standpoint. Since the training error has a downward bias and K-fold cross-validation has an upward bias, there will be an appropriate estimate in a family that connects the two estimates. In this paper, we investigate two families that connect the training error and K-fold cross-validation.

Traffic Flow Forecasting for Urban Work Zones None of numerous existing traffic flow forecasting models focus on work zones. Work zone events create conditions that are different from both normal operating conditions and incident conditions. In this paper, four models were developed for forecasting traffic flow for planned work zone events. The four models are random forest, regression tree, multilayer feedforward neural network, and nonparametric regression. Both long-term and short-term traffic flow forecasting applications were investigated. Long-term forecast involves forecasting 24 h in advance using historical traffic data, and short-term forecasts involves forecasting 1 h and 45, 30, and 15 min in advance using real-time temporal and spatial traffic data. Models were evaluated using data from work zone events on two types of roadways, a freeway, i.e., I-270, and a signalized arterial, i.e., MO-141, in St. Louis, MO, USA. The results showed that the random forest model yielded the most accurate long-term and short-term work zone traffic flow forecasts. For freeway data, the most influential variables were the latest interval's look-back traffic flows at the upstream, downstream, and current locations. For arterial data, the most influential variables were the traffic flows from the three look-back intervals at the current location only.

Daily long-term traffic flow forecasting based on a deep neural network. •A new deep learning algorithm to predict daily long-term traffic flow data using contextual factors.•Deep neutral network to mine the relationship between traffic flow data and contextual factors.•Advanced batch training can effectively improve convergence of the training process.

Cooperative Traffic Signal Control with Traffic Flow Prediction in Multi-Intersection. As traffic congestion in cities becomes serious, intelligent traffic signal control has been actively studied. Deep Q-Network (DQN), a representative deep reinforcement learning algorithm, is applied to various domains from fully-observable game environment to traffic signal control. Due to the effective performance of DQN, deep reinforcement learning has improved speeds and various DQN extensions have been introduced. However, most traffic signal control researches were performed at a single intersection, and because of the use of virtual simulators, there are limitations that do not take into account variables that affect actual traffic conditions. In this paper, we propose a cooperative traffic signal control with traffic flow prediction (TFP-CTSC) for a multi-intersection. A traffic flow prediction model predicts future traffic state and considers the variables that affect actual traffic conditions. In addition, for cooperative traffic signal control in multi-intersection, each intersection is modeled as an agent, and each agent is trained to take best action by receiving traffic states from the road environment. To deal with multi-intersection efficiently, agents share their traffic information with other adjacent intersections. In the experiment, TFP-CTSC is compared with existing traffic signal control algorithms in a 4 x 4 intersection environment. We verify our traffic flow prediction and cooperative method.

Weather Adaptive Traffic Prediction Using Neurowavelet Models Climate change is a prevalent issue facing the world today. Unexpected increase in rainfall intensity and events is one of the major signatures of climate change. Rainfall influences traffic conditions and, in turn, traffic volume in urban arterials. For improved traffic management under adverse weather conditions, it is important to develop a traffic prediction algorithm considering the effect of rainfall. This inclusion is not intuitive as the effect is not immediate, and the influence of rainfall on traffic volume is often unrecognizable in a direct correlation analysis between the two time-series data sets; it can only be observed at certain frequency levels. Accordingly, it is useful to employ a multiresolution prediction framework to develop a weather adaptive traffic forecasting algorithm. Discrete wavelet transform (DWT) is a well-known multiresolution data analysis methodology. However, DWT imparts time variance in the transformed signal and makes it unsuitable for further time-series analysis. Therefore, the stationary form of DWT known as stationary wavelet transform (SWT) has been used in this paper to develop a neurowavelet prediction algorithm to forecast hourly traffic flow considering the effect of rainfall. The proposed prediction algorithm has been evaluated at two urban arterial locations in Dublin, Ireland. This paper shows that the rainfall data successfully augments the traffic flow data as an exogenous variable in periods of inclement weather, resulting in accurate predictions of future traffic flow at the two chosen locations. The forecasts from the neurowavelet model outperform the forecasts from the standard artificial neural network (ANN) model.

An aggregation approach to short-term traffic flow prediction In this paper, an aggregation approach is proposed for traffic flow prediction that is based on the moving average (MA), exponential smoothing (ES), autoregressive MA (ARIMA), and neural network (NN) models. The aggregation approach assembles information from relevant time series. The source time series is the traffic flow volume that is collected 24 h/day over several years. The three relevant time series are a weekly similarity time series, a daily similarity time series, and an hourly time series, which can be directly generated from the source time series. The MA, ES, and ARIMA models are selected to give predictions of the three relevant time series. The predictions that result from the different models are used as the basis of the NN in the aggregation stage. The output of the trained NN serves as the final prediction. To assess the performance of the different models, the naïve, ARIMA, nonparametric regression, NN, and data aggregation (DA) models are applied to the prediction of a real vehicle traffic flow, from which data have been collected at a data-collection point that is located on National Highway 107, Guangzhou, Guangdong, China. The outcome suggests that the DA model obtains a more accurate forecast than any individual model alone. The aggregation strategy can offer substantial benefits in terms of improving operational forecasting.

Explanations and Expectations: Trust Building in Automated Vehicles. Trust is a vital determinant of acceptance of automated vehicles (AVs) and expectations and explanations are often at the heart of any trusting relationship. Once expectations have been violated, explanations are needed to mitigate the damage. This study introduces the importance of timing of explanations in promoting trust in AVs. We present the preliminary results of a within-subjects experimental study involving eight participants exposed to four AV driving conditions (i.e. 32 data points). Preliminary results show a pattern that suggests that explanations provided before the AV takes actions promote more trust than explanations provided afterward.

Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation Object detection performance, as measured on the canonical PASCAL VOC dataset, has plateaued in the last few years. The best-performing methods are complex ensemble systems that typically combine multiple low-level image features with high-level context. In this paper, we propose a simple and scalable detection algorithm that improves mean average precision (mAP) by more than 30% relative to the previous best result on VOC 2012 -- achieving a mAP of 53.3%. Our approach combines two key insights: (1) one can apply high-capacity convolutional neural networks (CNNs) to bottom-up region proposals in order to localize and segment objects and (2) when labeled training data is scarce, supervised pre-training for an auxiliary task, followed by domain-specific fine-tuning, yields a significant performance boost. Since we combine region proposals with CNNs, we call our method R-CNN: Regions with CNN features. We also present experiments that provide insight into what the network learns, revealing a rich hierarchy of image features. Source code for the complete system is available at http://www.cs.berkeley.edu/~rbg/rcnn.

Certifying and Removing Disparate Impact What does it mean for an algorithm to be biased? In U.S. law, unintentional bias is encoded via disparate impact, which occurs when a selection process has widely different outcomes for different groups, even as it appears to be neutral. This legal determination hinges on a definition of a protected class (ethnicity, gender) and an explicit description of the process. When computers are involved, determining disparate impact (and hence bias) is harder. It might not be possible to disclose the process. In addition, even if the process is open, it might be hard to elucidate in a legal setting how the algorithm makes its decisions. Instead of requiring access to the process, we propose making inferences based on the data it uses. We present four contributions. First, we link disparate impact to a measure of classification accuracy that while known, has received relatively little attention. Second, we propose a test for disparate impact based on how well the protected class can be predicted from the other attributes. Third, we describe methods by which data might be made unbiased. Finally, we present empirical evidence supporting the effectiveness of our test for disparate impact and our approach for both masking bias and preserving relevant information in the data. Interestingly, our approach resembles some actual selection practices that have recently received legal scrutiny.

Ambulatory measurement of shoulder and elbow kinematics through inertial and magnetic sensors. Inertial and magnetic measurement systems (IMMSs) are a new generation of motion analysis systems which may diffuse the measurement of upper-limb kinematics to ambulatory settings. Based on the MT9B IMMS (Xsens Technologies, NL), we therefore developed a protocol that measures the scapulothoracic, humerothoracic and elbow 3D kinematics. To preliminarily evaluate the protocol, a 23-year-old subject performed six tasks involving shoulder and elbow single-joint-angle movements. Criteria for protocol validity were limited cross-talk with the other joint-angles during each task; scapulohumeral-rhythm close to literature results; and constant carrying-angle. To assess the accuracy of the MT9B when measuring the upper-limb kinematics through the protocol, we compared the MT9B estimations during the six tasks, plus other four, with the estimations of an optoelectronic system (the gold standard), in terms of RMS error, correlation coefficient (r), and the amplitude ratio (m). Results indicate that the criteria for protocol validity were met for all tasks. For the joint angles mainly involved in each movement, the MT9B estimations presented RMS errors r > 0.99 and 0.9 m < 1.09. It appears therefore that (1) the protocol in combination with the MT9B is valid for, and (2) the MT9B in combination with the protocol is accurate when, measuring shoulder and elbow kinematics, during the tasks tested, in ambulatory settings.

Dynamic Trajectory Planning for Vehicle Autonomous Driving Trajectory planning is one of the key and challenging tasks in autonomous driving. This paper proposes a novel method that dynamically plans trajectories, with the aim to achieve quick and safe reaction to the changing driving environment and optimal balance between vehicle performance and driving comfort. With the proposed method, such complex maneuvers can be decomposed into two sub-maneuvers, i.e., lane change and lane keeping, or their combinations, such that the trajectory planning is generalized and simplified, mainly based on lane change maneuvers. A two fold optimization-based method is proposed for stationary trajectory planning as well as dynamic trajectory planning in the presence of a dynamic traffic environment. Simulation is conducted to demonstrate the efficiency and effectiveness of the proposed method.

Wireless Powered Sensor Networks for Internet of Things: Maximum Throughput and Optimal Power Allocation. This paper investigates a wireless powered sensor network, where multiple sensor nodes are deployed to monitor a certain external environment. A multiantenna power station (PS) provides the power to these sensor nodes during wireless energy transfer phase, and consequently the sensor nodes employ the harvested energy to transmit their own monitoring information to a fusion center during wireless i...

Attitudes Towards Social Robots In Education: Enthusiast, Practical, Troubled, Sceptic, And Mindfully Positive While social robots bring new opportunities for education, they also come with moral challenges. Therefore, there is a need for moral guidelines for the responsible implementation of these robots. When developing such guidelines, it is important to include different stakeholder perspectives. Existing (qualitative) studies regarding these perspectives however mainly focus on single stakeholders. In this exploratory study, we examine and compare the attitudes of multiple stakeholders on the use of social robots in primary education, using a novel questionnaire that covers various aspects of moral issues mentioned in earlier studies. Furthermore, we also group the stakeholders based on similarities in attitudes and examine which socio-demographic characteristics influence these attitude types. Based on the results, we identify five distinct attitude profiles and show that the probability of belonging to a specific profile is affected by such characteristics as stakeholder type, age, education and income. Our results also indicate that social robots have the potential to be implemented in education in a morally responsible way that takes into account the attitudes of various stakeholders, although there are multiple moral issues that need to be addressed first. Finally, we present seven (practical) implications for a responsible application of social robots in education following from our results. These implications provide valuable insights into how social robots should be implemented.

Biometrics: Trust, But Verify Over the past two decades, biometric recognition has exploded into a plethora of different applications around the globe. This proliferation can be attributed to the high levels of authentication accuracy and user convenience that biometric recognition systems afford end-users. However, in-spite of the success of biometric recognition systems, there are a number of outstanding problems and concerns pertaining to the various sub-modules of biometric recognition systems that create an element of mistrust in their use - both by the scientific community and also the public at large. Some of these problems include: i) questions related to system recognition performance, ii) security (spoof attacks, adversarial attacks, template reconstruction attacks and demographic information leakage), iii) uncertainty over the bias and fairness of the systems to all users, iv) explainability of the seemingly black-box decisions made by most recognition systems, and v) concerns over data centralization and user privacy. In this paper, we provide an overview of each of the aforementioned open-ended challenges. We survey work that has been conducted to address each of these concerns and highlight the issues requiring further attention. Finally, we provide insights into how the biometric community can address core biometric recognition systems design issues to better instill trust, fairness, and security for all.

The Sybil Attack Large-scale peer-to-peer systems facesecurity threats from faulty or hostile remotecomputing elements. To resist these threats, manysuch systems employ redundancy. However, if asingle faulty entity can present multiple identities,it can control a substantial fraction of the system,thereby undermining this redundancy. Oneapproach to preventing these &quot;Sybil attacks&quot; is tohave a trusted agency certify identities. Thispaper shows that, without a logically centralizedauthority, Sybil...

BLEU: a method for automatic evaluation of machine translation Human evaluations of machine translation are extensive but expensive. Human evaluations can take months to finish and involve human labor that can not be reused. We propose a method of automatic machine translation evaluation that is quick, inexpensive, and language-independent, that correlates highly with human evaluation, and that has little marginal cost per run. We present this method as an automated understudy to skilled human judges which substitutes for them when there is need for quick or frequent evaluations.

Computational thinking Summary form only given. My vision for the 21st century, Computational Thinking, will be a fundamental skill used by everyone in the world. To reading, writing, and arithmetic, we should add computational thinking to every child's analytical ability. Computational thinking involves solving problems, designing systems, and understanding human behavior by drawing on the concepts fundamental to computer science. Thinking like a computer scientist means more than being able to program a computer. It requires the ability to abstract and thus to think at multiple levels of abstraction. In this talk I will give many examples of computational thinking, argue that it has already influenced other disciplines, and promote the idea that teaching computational thinking can not only inspire future generations to enter the field of computer science but benefit people in all fields.

Fuzzy logic in control systems: fuzzy logic controller. I.

Switching between stabilizing controllers This paper deals with the problem of switching between several linear time-invariant (LTI) controllers—all of them capable of stabilizing a speci4c LTI process—in such a way that the stability of the closed-loop system is guaranteed for any switching sequence. We show that it is possible to 4nd realizations for any given family of controller transfer matrices so that the closed-loop system remains stable, no matter how we switch among the controller. The motivation for this problem is the control of complex systems where con8icting requirements make a single LTI controller unsuitable. ? 2002 Published by Elsevier Science Ltd.

Tabu Search - Part I

Bidirectional recurrent neural networks In the first part of this paper, a regular recurrent neural network (RNN) is extended to a bidirectional recurrent neural network (BRNN). The BRNN can be trained without the limitation of using input information just up to a preset future frame. This is accomplished by training it simultaneously in positive and negative time direction. Structure and training procedure of the proposed network are explained. In regression and classification experiments on artificial data, the proposed structure gives better results than other approaches. For real data, classification experiments for phonemes from the TIMIT database show the same tendency. In the second part of this paper, it is shown how the proposed bidirectional structure can be easily modified to allow efficient estimation of the conditional posterior probability of complete symbol sequences without making any explicit assumption about the shape of the distribution. For this part, experiments on real data are reported

An intensive survey of fair non-repudiation protocols With the phenomenal growth of the Internet and open networks in general, security services, such as non-repudiation, become crucial to many applications. Non-repudiation services must ensure that when Alice sends some information to Bob over a network, neither Alice nor Bob can deny having participated in a part or the whole of this communication. Therefore a fair non-repudiation protocol has to generate non-repudiation of origin evidences intended to Bob, and non-repudiation of receipt evidences destined to Alice. In this paper, we clearly define the properties a fair non-repudiation protocol must respect, and give a survey of the most important non-repudiation protocols without and with trusted third party (TTP). For the later ones we discuss the evolution of the TTP's involvement and, between others, describe the most recent protocol using a transparent TTP. We also discuss some ad-hoc problems related to the management of non-repudiation evidences.

Dynamic movement and positioning of embodied agents in multiparty conversations For embodied agents to engage in realistic multiparty conversation, they must stand in appropriate places with respect to other agents and the environment. When these factors change, such as an agent joining the conversation, the agents must dynamically move to a new location and/or orientation to accommodate. This paper presents an algorithm for simulating movement of agents based on observed human behavior using techniques developed for pedestrian movement in crowd simulations. We extend a previous group conversation simulation to include an agent motion algorithm. We examine several test cases and show how the simulation generates results that mirror real-life conversation settings.

An improved genetic algorithm with conditional genetic operators and its application to set-covering problem The genetic algorithm (GA) is a popular, biologically inspired optimization method. However, in the GA there is no rule of thumb to design the GA operators and select GA parameters. Instead, trial-and-error has to be applied. In this paper we present an improved genetic algorithm in which crossover and mutation are performed conditionally instead of probability. Because there are no crossover rate and mutation rate to be selected, the proposed improved GA can be more easily applied to a problem than the conventional genetic algorithms. The proposed improved genetic algorithm is applied to solve the set-covering problem. Experimental studies show that the improved GA produces better results over the conventional one and other methods.

Lane-level traffic estimations using microscopic traffic variables This paper proposes a novel inference method to estimate lane-level traffic flow, time occupancy and vehicle inter-arrival time on road segments where local information could not be measured and assessed directly. The main contributions of the proposed method are 1) the ability to perform lane-level estimations of traffic flow, time occupancy and vehicle inter-arrival time and 2) the ability to adapt to different traffic regimes by assessing only microscopic traffic variables. We propose a modified Kriging estimation model which explicitly takes into account both spatial and temporal variability. Performance evaluations are conducted using real-world data under different traffic regimes and it is shown that the proposed method outperforms a Kalman filter-based approach.

Convolutional Neural Network-Based Classification of Driver's Emotion during Aggressive and Smooth Driving Using Multi-Modal Camera Sensors. Because aggressive driving often causes large-scale loss of life and property, techniques for advance detection of adverse driver emotional states have become important for the prevention of aggressive driving behaviors. Previous studies have primarily focused on systems for detecting aggressive driver emotion via smart-phone accelerometers and gyro-sensors, or they focused on methods of detecting physiological signals using electroencephalography (EEG) or electrocardiogram (ECG) sensors. Because EEG and ECG sensors cause discomfort to drivers and can be detached from the driver's body, it becomes difficult to focus on bio-signals to determine their emotional state. Gyro-sensors and accelerometers depend on the performance of GPS receivers and cannot be used in areas where GPS signals are blocked. Moreover, if driving on a mountain road with many quick turns, a driver's emotional state can easily be misrecognized as that of an aggressive driver. To resolve these problems, we propose a convolutional neural network (CNN)-based method of detecting emotion to identify aggressive driving using input images of the driver's face, obtained using near-infrared (NIR) light and thermal camera sensors. In this research, we conducted an experiment using our own database, which provides a high classification accuracy for detecting driver emotion leading to either aggressive or smooth (i.e., relaxed) driving. Our proposed method demonstrates better performance than existing methods.

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

Open social student modeling: visualizing student models with parallel introspectiveviews This paper explores a social extension of open student modeling that we call open social student modeling. We present a specific implementation of this approach that uses parallel IntrospectiveViews to visualize models representing student progress with QuizJET parameterized self-assessment questions for Java programming. The interface allows visualizing not only the student's own model, but also displaying parallel views on the models of their peers and the cumulative model of the entire class or group. The system was evaluated in a semester-long classroom study. While the use of the system was non-mandatory, the parallel IntrospectiveViews interface caused an increase in all of the usage parameters in comparison to a regular portal-based access, which allowed the student to achieve a higher success rate in answering the questions. The collected data offer some evidence that a combination of traditional personalized guidance with social guidance was more effective than personalized guidance alone.

Lifelong Learner Modeling for Lifelong Personalized Pervasive Learning Pervasive and ubiquitous computing have the potential to make huge changes in the ways that we will learn, throughout our lives. This paper presents a vision for the lifelong user model as a first class citizen, existing independently of any single application and controlled by the learner. The paper argues that this is a critical foundation for a vision of personalised lifelong learning as well as a form of augmented cognition that enables learners to supplement their own knowledge with readily accessible digital information based on documents that they have accessed or used. The paper presents work that provides foundations for this vision for a lifelong user model. First, it outlines technical issues and research into approaches for addressing them. Then it presents work on the interface between the learner and the lifelong user model because the human issues of control and privacy are so central. The final discussion and conclusions draw upon these to define a roadmap for future research in a selection of the key areas that will underpin this vision of the lifelong user model.

AnnotatEd: A social navigation and annotation service for web-based educational resources Web page annotation and adaptive navigation support are two active, but independent research directions focused on the same goal: expanding the functionality of the Web as a hypertext system. The goal of the AnnotatEd system presented in this paper has been to integrate annotation and adaptive navigation support into a single value-added service where the components can reinforce each other and create new unique attributes. This paper describes the implementation of AnnotatEd from early prototypes to the current version, which has been explored in several contexts. We summarize some lessons we learned during the development process and which defined the current functionality of the system. We also present the results of several classroom studies of the system. These results demonstrate the importance of the browsing-based information access supported by AnnotatEd and the value of both the annotation and navigation support functionalities offered by the system.

A Web-Based Tool For Control Engineering Teaching In this article a new tool for control engineering teaching is presented. The tool was implemented using Java applets and is freely accessible through Web. It allows the analysis and simulation of linear control systems and was created to complement the theoretical lectures in basic control engineering courses. The article is not only centered in the description of the tool but also in the methodology to use it and its evaluation in an electrical engineering degree. Two practical problems are included in the manuscript to illustrate the use of the main functions implemented. The developed web-based tool can be accessed through the link http://www.controlweb.cyc.ull.es. (C) 2006 Wiley Periodicals, Inc.

Active Open Learner Models as Animal Companions: Motivating Children to Learn through Interacting with My-Pet and Our-Pet This pilot study reports how to portray open learner models as animal companions in order to motivate children to learn in the digital classroom environment. To meet two challenges of motivation and interactivity for open learner models, the concept of open learner models as animal companions is proposed based on the emotional attachment of humans towards pets. Animal companions adopt three strategies and play various educational roles to help children's learning in motivation, reflection and member interactions. A class of students is divided into several teams. A student keeps her own individual animal companion, called My-Pet, which holds the open learner model of the student, and each team has a team animal companion, called Our-Pet, which owns their open group learner model. A preliminary experiment is conducted in a fifth-grade class with 31 eleven year old students in an elementary school to collect initial feedback in cognitive and affective aspects.

Guiding students to the right questions: adaptive navigation support in an E-Learning system for Java programming Rapid growth of the volume of interactive questions available to the students of modern E-Learning courses placed the problem of personalized guidance on the agenda of E-Learning researchers. Without proper guidance, students frequently select too simple or too complicated problems and ended either bored or discouraged. This paper explores a specific personalized guidance technology known as adaptive navigation support. We developed JavaGuide, a system, which guides students to appropriate questions in a Java programming course, and investigated the effect of personalized guidance a three-semester long classroom study. The results of this study confirm the educational and motivational effects of adaptive navigation support.

Using Social Psychology to Motivate Contributions to Online Communities Under-contribution is a problem for many online communities. Social psychology theories of social loafing and goal-setting can lead to mid-level design goals to address this problem. We tested design principles derived from these theories in four field experiments involving members of an online movie recommender community. In each of the experiments participated were given different explanations for the value of their contributions. As predicted by theory, individuals contributed when they were reminded of their uniqueness and when they were given specific and challenging goals. However, other predictions were disconfirmed. For example, in one experiment, participants given group goals contributed more than those given individual goals. The article ends with suggestions and challenges for mining design implications from social science theories.

On the security of public key protocols Recently the use of public key encryption to provide secure network communication has received considerable attention. Such public key systems are usually effective against passive eavesdroppers, who merely tap the lines and try to decipher the message. It has been pointed out, however, that an improperly designed protocol could be vulnerable to an active saboteur, one who may impersonate another user or alter the message being transmitted. Several models are formulated in which the security of protocols can be discussed precisely. Algorithms and characterizations that can be used to determine protocol security in these models are given.

Multi-Hop Cooperative Computation Offloading for Industrial IoT–Edge–Cloud Computing Environments The concept of the industrial Internet of things (IIoT) is being widely applied to service provisioning in many domains, including smart healthcare, intelligent transportation, autopilot, and the smart grid. However, because of the IIoT devices’ limited onboard resources, supporting resource-intensive applications, such as 3D sensing, navigation, AI processing, and big-data analytics, remains a challenging task. In this paper, we study the multi-hop computation-offloading problem for the IIoT–edge–cloud computing model and adopt a game-theoretic approach to achieving Quality of service (QoS)-aware computation offloading in a distributed manner. First, we study the computation-offloading and communication-routing problems with the goal of minimizing each task's computation time and energy consumption, formulating the joint problem as a potential game in which the IIoT devices determine their computation-offloading strategies. Second, we apply a free–bound mechanism that can ensure a finite improvement path to a Nash equilibrium. Third, we propose a multi-hop cooperative-messaging mechanism and develop two QoS-aware distributed algorithms that can achieve the Nash equilibrium. Our simulation results show that our algorithms offer a stable performance gain for IIoT in various scenarios and scale well as the device size increases.

A hierarchical type-2 fuzzy logic control architecture for autonomous mobile robots Autonomous mobile robots navigating in changing and dynamic unstructured environments like the outdoor environments need to cope with large amounts of uncertainties that are inherent of natural environments. The traditional type-1 fuzzy logic controller (FLC) using precise type-1 fuzzy sets cannot fully handle such uncertainties. A type-2 FLC using type-2 fuzzy sets can handle such uncertainties to produce a better performance. In this paper, we present a novel reactive control architecture for autonomous mobile robots that is based on type-2 FLC to implement the basic navigation behaviors and the coordination between these behaviors to produce a type-2 hierarchical FLC. In our experiments, we implemented this type-2 architecture in different types of mobile robots navigating in indoor and outdoor unstructured and challenging environments. The type-2-based control system dealt with the uncertainties facing mobile robots in unstructured environments and resulted in a very good performance that outperformed the type-1-based control system while achieving a significant rule reduction compared to the type-1 system.

State resetting for bumpless switching in supervisory control In this paper the realization and implementation of a multi-controller scheme made of a finite set of linear single-input-single-output controllers, possibly having different state dimensions, is studied. The supervisory control framework is considered, namely a minimal parameter dependent realization of the set of controllers such that all controllers share the same state space is used. A specific state resetting strategy based on the behavioral approach to system theory is developed in order to master the transient upon controller switching.

Beamforming for MISO Interference Channels with QoS and RF Energy Transfer We consider a multiuser multiple-input single-output interference channel where the receivers are characterized by both quality-of-service (QoS) and radio-frequency (RF) energy harvesting (EH) constraints. We consider the power splitting RF-EH technique where each receiver divides the received signal into two parts a) for information decoding and b) for battery charging. The minimum required power that supports both the QoS and the RF-EH constraints is formulated as an optimization problem that incorporates the transmitted power and the beamforming design at each transmitter as well as the power splitting ratio at each receiver. We consider both the cases of fixed beamforming and when the beamforming design is incorporated into the optimization problem. For fixed beamforming we study three standard beamforming schemes, the zero-forcing (ZF), the regularized zero-forcing (RZF) and the maximum ratio transmission (MRT); a hybrid scheme, MRT-ZF, comprised of a linear combination of MRT and ZF beamforming is also examined. The optimal solution for ZF beamforming is derived in closed-form, while optimization algorithms based on second-order cone programming are developed for MRT, RZF and MRT-ZF beamforming to solve the problem. In addition, the joint-optimization of beamforming and power allocation is studied using semidefinite programming (SDP) with the aid of rank relaxation.

Collective feature selection to identify crucial epistatic variants. In this study, we were able to show that selecting variables using a collective feature selection approach could help in selecting true positive epistatic variables more frequently than applying any single method for feature selection via simulation studies. We were able to demonstrate the effectiveness of collective feature selection along with a comparison of many methods in our simulation analysis. We also applied our method to identify non-linear networks associated with obesity.

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

A parallel grid-search-based SVM optimization algorithm on Spark for passenger hotspot prediction Predicting passenger hotspots helps drivers quickly pick up travelers, reduces cruise expenses, and maximizes revenue per unit time in intelligent transportation systems. To improve the accuracy and robustness of passenger hotspot prediction (PHP), this paper proposes a parallel Grid-Search-based Support Vector Machine (GS-SVM) optimization algorithm on Spark, which provides an efficient methodology to search for passengers in a complex urban traffic network quickly. Specifically, to effectively locate passenger hotspots, an urban road network is gridded on the Spark parallel distributed computing platform. Moreover, to enhance the accuracy of PHP, the grid search (GS) approach is employed to optimize the radial basis function (RBF) of the support vector machine (SVM), and the cross-validation method is utilized to find out the global optimal parameter combination. Finally, the SVM optimization algorithm is implemented on Spark to improve the robustness of PHP. In particular, the proposed GS-SVM algorithm is applied to successfully predict passenger hotspots. By analyzing seven groups of data sets and comparing with serval state-of-the-art algorithms including autoregressive integrated moving average (ARIMA), support vector regression (SVR), long short-term memory (LSTM), and convolutional neural network (CNN), the results of an empirical study indicate that the MAPE value of our GS-SVM algorithm is lower than that of comparative algorithms at least 78.4%.

The Sybil Attack Large-scale peer-to-peer systems facesecurity threats from faulty or hostile remotecomputing elements. To resist these threats, manysuch systems employ redundancy. However, if asingle faulty entity can present multiple identities,it can control a substantial fraction of the system,thereby undermining this redundancy. Oneapproach to preventing these &quot;Sybil attacks&quot; is tohave a trusted agency certify identities. Thispaper shows that, without a logically centralizedauthority, Sybil...

BLEU: a method for automatic evaluation of machine translation Human evaluations of machine translation are extensive but expensive. Human evaluations can take months to finish and involve human labor that can not be reused. We propose a method of automatic machine translation evaluation that is quick, inexpensive, and language-independent, that correlates highly with human evaluation, and that has little marginal cost per run. We present this method as an automated understudy to skilled human judges which substitutes for them when there is need for quick or frequent evaluations.

Computational thinking Summary form only given. My vision for the 21st century, Computational Thinking, will be a fundamental skill used by everyone in the world. To reading, writing, and arithmetic, we should add computational thinking to every child's analytical ability. Computational thinking involves solving problems, designing systems, and understanding human behavior by drawing on the concepts fundamental to computer science. Thinking like a computer scientist means more than being able to program a computer. It requires the ability to abstract and thus to think at multiple levels of abstraction. In this talk I will give many examples of computational thinking, argue that it has already influenced other disciplines, and promote the idea that teaching computational thinking can not only inspire future generations to enter the field of computer science but benefit people in all fields.

Fuzzy logic in control systems: fuzzy logic controller. I.

Switching between stabilizing controllers This paper deals with the problem of switching between several linear time-invariant (LTI) controllers—all of them capable of stabilizing a speci4c LTI process—in such a way that the stability of the closed-loop system is guaranteed for any switching sequence. We show that it is possible to 4nd realizations for any given family of controller transfer matrices so that the closed-loop system remains stable, no matter how we switch among the controller. The motivation for this problem is the control of complex systems where con8icting requirements make a single LTI controller unsuitable. ? 2002 Published by Elsevier Science Ltd.

Tabu Search - Part I

Bidirectional recurrent neural networks In the first part of this paper, a regular recurrent neural network (RNN) is extended to a bidirectional recurrent neural network (BRNN). The BRNN can be trained without the limitation of using input information just up to a preset future frame. This is accomplished by training it simultaneously in positive and negative time direction. Structure and training procedure of the proposed network are explained. In regression and classification experiments on artificial data, the proposed structure gives better results than other approaches. For real data, classification experiments for phonemes from the TIMIT database show the same tendency. In the second part of this paper, it is shown how the proposed bidirectional structure can be easily modified to allow efficient estimation of the conditional posterior probability of complete symbol sequences without making any explicit assumption about the shape of the distribution. For this part, experiments on real data are reported

An intensive survey of fair non-repudiation protocols With the phenomenal growth of the Internet and open networks in general, security services, such as non-repudiation, become crucial to many applications. Non-repudiation services must ensure that when Alice sends some information to Bob over a network, neither Alice nor Bob can deny having participated in a part or the whole of this communication. Therefore a fair non-repudiation protocol has to generate non-repudiation of origin evidences intended to Bob, and non-repudiation of receipt evidences destined to Alice. In this paper, we clearly define the properties a fair non-repudiation protocol must respect, and give a survey of the most important non-repudiation protocols without and with trusted third party (TTP). For the later ones we discuss the evolution of the TTP's involvement and, between others, describe the most recent protocol using a transparent TTP. We also discuss some ad-hoc problems related to the management of non-repudiation evidences.

Dynamic movement and positioning of embodied agents in multiparty conversations For embodied agents to engage in realistic multiparty conversation, they must stand in appropriate places with respect to other agents and the environment. When these factors change, such as an agent joining the conversation, the agents must dynamically move to a new location and/or orientation to accommodate. This paper presents an algorithm for simulating movement of agents based on observed human behavior using techniques developed for pedestrian movement in crowd simulations. We extend a previous group conversation simulation to include an agent motion algorithm. We examine several test cases and show how the simulation generates results that mirror real-life conversation settings.

An improved genetic algorithm with conditional genetic operators and its application to set-covering problem The genetic algorithm (GA) is a popular, biologically inspired optimization method. However, in the GA there is no rule of thumb to design the GA operators and select GA parameters. Instead, trial-and-error has to be applied. In this paper we present an improved genetic algorithm in which crossover and mutation are performed conditionally instead of probability. Because there are no crossover rate and mutation rate to be selected, the proposed improved GA can be more easily applied to a problem than the conventional genetic algorithms. The proposed improved genetic algorithm is applied to solve the set-covering problem. Experimental studies show that the improved GA produces better results over the conventional one and other methods.

Lane-level traffic estimations using microscopic traffic variables This paper proposes a novel inference method to estimate lane-level traffic flow, time occupancy and vehicle inter-arrival time on road segments where local information could not be measured and assessed directly. The main contributions of the proposed method are 1) the ability to perform lane-level estimations of traffic flow, time occupancy and vehicle inter-arrival time and 2) the ability to adapt to different traffic regimes by assessing only microscopic traffic variables. We propose a modified Kriging estimation model which explicitly takes into account both spatial and temporal variability. Performance evaluations are conducted using real-world data under different traffic regimes and it is shown that the proposed method outperforms a Kalman filter-based approach.

Convolutional Neural Network-Based Classification of Driver's Emotion during Aggressive and Smooth Driving Using Multi-Modal Camera Sensors. Because aggressive driving often causes large-scale loss of life and property, techniques for advance detection of adverse driver emotional states have become important for the prevention of aggressive driving behaviors. Previous studies have primarily focused on systems for detecting aggressive driver emotion via smart-phone accelerometers and gyro-sensors, or they focused on methods of detecting physiological signals using electroencephalography (EEG) or electrocardiogram (ECG) sensors. Because EEG and ECG sensors cause discomfort to drivers and can be detached from the driver's body, it becomes difficult to focus on bio-signals to determine their emotional state. Gyro-sensors and accelerometers depend on the performance of GPS receivers and cannot be used in areas where GPS signals are blocked. Moreover, if driving on a mountain road with many quick turns, a driver's emotional state can easily be misrecognized as that of an aggressive driver. To resolve these problems, we propose a convolutional neural network (CNN)-based method of detecting emotion to identify aggressive driving using input images of the driver's face, obtained using near-infrared (NIR) light and thermal camera sensors. In this research, we conducted an experiment using our own database, which provides a high classification accuracy for detecting driver emotion leading to either aggressive or smooth (i.e., relaxed) driving. Our proposed method demonstrates better performance than existing methods.

Ethical Considerations Of Applying Robots In Kindergarten Settings: Towards An Approach From A Macroperspective In child-robot interaction (cHRI) research, many studies pursue the goal to develop interactive systems that can be applied in everyday settings. For early education, increasingly, the setting of a kindergarten is targeted. However, when cHRI and research are brought into a kindergarten, a range of ethical and related procedural aspects have to be considered and dealt with. While ethical models elaborated within other human-robot interaction settings, e.g., assisted living contexts, can provide some important indicators for relevant issues, we argue that it is important to start developing a systematic approach to identify and tackle those ethical issues which rise with cHRI in kindergarten settings on a more global level and address the impact of the technology from a macroperspective beyond the effects on the individual. Based on our experience in conducting studies with children in general and pedagogical considerations on the role of the institution of kindergarten in specific, in this paper, we enfold some relevant aspects that have barely been addressed in an explicit way in current cHRI research. Four areas are analyzed and key ethical issues are identified in each area: (1) the institutional setting of a kindergarten, (2) children as a vulnerable group, (3) the caregivers' role, and (4) pedagogical concepts. With our considerations, we aim at (i) broadening the methodology of the current studies within the area of cHRI, (ii) revalidate it based on our comprehensive empirical experience with research in kindergarten settings, both laboratory and real-world contexts, and (iii) provide a framework for the development of a more systematic approach to address the ethical issues in cHRI research within kindergarten settings.

A Secure Framework For Remote Diagnosis In Health Care: A High Capacity Reversible Data Hiding Technique For Medical Images The health care industry involves the processing of large-scale images for various applications. Remote diagnosis is one such significant area where medical images are sent across vulnerable communication media. Hence, a secure and robust framework is necessary to hide and retrieve patient information in medical images. Here, we propose a high capacity reversible data hiding technique that can be used to embed patient data using a new weighted interpolation technique. In this approach, to improve the payload capacity, interpolated pixels are effectively utilized for the data embedding process using modular arithmetic. The original cover pixels are also employed to embed data using the difference expansion method. The framework developed is tested with standard benchmark images and medical images. The experimental results prove that the proposed method proffers a better output in comparison with the other contemporary methods in this domain.

On (k, n)*-visual cryptography scheme Let P = {1, 2, . . . , n} be a set of elements called participants. In this paper we construct a visual cryptography scheme (VCS) for the strong access structure specified by the set Γ0 of all minimal qualified sets, where $${\Gamma_0=\{S: S\subseteq P, 1\in S}$$ and |S| = k}. Any VCS for this strong access structure is called a (k, n)*-VCS. We also obtain bounds for the optimal pixel expansion and optimal relative contrast for a (k, n)*-VCS.

Selective bit embedding scheme for robust blind color image watermarking. •A novel blind color image watermarking that uses gray-scale image as watermark.•An efficient embedding rule for wavelet coefficient difference quantization.•2D Otsu algorithm for high accuracy of watermark recovery.•A good performance balance of triangular imperceptibility-robustness-payload.•Outperformance of robustness to state-of-the-art approaches at high payload capacity.

Optimized Gabor Feature Extraction for Mass Classification Using Cuckoo Search for Big Data E-Healthcare Widespread use of electronic health records is a major cause of a massive dataset that ultimately results in Big Data. Computer-aided systems for healthcare can be an effective tool to automatically process such big data. Breast cancer is one of the major causes of high mortality rate among women in the world since it is difficult to detect due to lack of early symptoms. There is a number of techniques and advanced technologies available to detect breast tumors nowadays. One of the common approaches for breast tumour detection is mammography. The similarity between the normal (unaffected) tissues and the masses (affected) tissues is often very high that leads to false positives (FP). In the field of medicine, the sensitivity to false positives is very high because it results in false diagnosis and can lead to serious consequences. Therefore, it is a challenge for the researchers to correctly distinguish between the normal and affected tissues to increase the detection accuracy. Radiologists use Gabor filter bank for feature extraction and apply it to the entire input image that yields poor results. The proposed system optimizes the Gabor filter bank to select most appropriate Gabor filter using a metaheuristic algorithm known as “Cuckoo Search”. The proposed algorithm is run over sub-images in order to extract more descriptive features. Moreover, feature subset selection is used to reduce feature size because feature extracted from the segmented region of interest will be high dimensional and cannot be handled easily. This algorithm is more efficient, fast, and less complex and spawns improved results. The proposed method is tested on 2000 mammograms taken from DDSM database and outperforms some of the best techniques used for mammogram classification based on Sensitivity, Specificity, Accuracy, and Area under the curve (ROC).

A New Method For Producing 320-Bit Modified Hash Towards Tamper Detection And Restoration In Colour Images Image sharing in open and public communication infrastructure is vulnerable. There are several solutions articulated by worldwide researchers for secure image transmission. However, with the increasing computing capacity and capability, the currently available security solutions are being breached. This yearns for more robust solutions. The significant contribution of the proposed work involves in the generation of a new hash algorithm for the entire RGB image of size 256 x 256 with no Region of Interest (RoI) restriction. This paper mainly focuses on integrity verification, tamper detection, localisation of the tampered area and its restoration. The whole image is considered as sensitive data for which one-way integrity verification code is generated block-wise to locate the areas of tampering. Integrity validation phase will compare the received digest and generated the digest from the received image. Blocks which are failed to pass in integrity validation will undergo the recovery process.

Novel robust zero-watermarking scheme for digital rights management of 3D videos. Digital watermarking is an effective technique for the digital rights management (DRM) of three-dimensional (3D) videos, which is still a crucial issue in the field of 3D televisions. The current watermarking schemes for 3D videos can be classified into two main categories: One embeds watermarks into 2D video frames, and the other embeds watermarks into depth maps. The former causes irreversible distortions to synthesized 3D videos whereas the latter is insufficiently robust against some of normal video attacks. Moreover, because these watermarking schemes only embed watermarks into either 2D video frames or depth maps of 3D videos, none of them can protect the copyrights of these two parts simultaneously and independently. To address those problems, a novel robust zero-watermarking scheme for the DRM of 3D videos is proposed in this study. In the proposed scheme, master shares are first generated by extracting features from temporally informative representative images (TIRIs) of both the 2D video frames and depth maps. Then, ownership shares, which denote the relationships between copyright information and master shares, are generated based on the Visual Secret Sharing (VSS) scheme and stored for copyright identification. Finally, copyright ownerships are identified by stacking master shares and ownership shares. The experiment results demonstrate that the proposed scheme outperforms the current state-of-the-art watermarking schemes because it does not cause any distortion to the synthesized 3D videos, exhibits strong robustness against various video attacks, and protects the copyrights of 2D video frames and depth maps of 3D videos simultaneously and independently. A robust lossless zero-watermarking scheme for the digital rights management (DRM) of DIBR-based 3D videos is proposed.The proposed scheme is based on Visual Secret Sharing and fully utilizes the robust content-based features of both 2D frames and depth maps of 3D videos.The copyrights of 2D video frames and depth maps of 3D videos are protected simultaneously and independently by designing a flexible copyright identification mechanism to fully satisfy the DRM requirements of different DIBR-based 3D videos.

A survey on ear biometrics Recognizing people by their ear has recently received significant attention in the literature. Several reasons account for this trend: first, ear recognition does not suffer from some problems associated with other non-contact biometrics, such as face recognition; second, it is the most promising candidate for combination with the face in the context of multi-pose face recognition; and third, the ear can be used for human recognition in surveillance videos where the face may be occluded completely or in part. Further, the ear appears to degrade little with age. Even though current ear detection and recognition systems have reached a certain level of maturity, their success is limited to controlled indoor conditions. In addition to variation in illumination, other open research problems include hair occlusion, earprint forensics, ear symmetry, ear classification, and ear individuality. This article provides a detailed survey of research conducted in ear detection and recognition. It provides an up-to-date review of the existing literature revealing the current state-of-art for not only those who are working in this area but also for those who might exploit this new approach. Furthermore, it offers insights into some unsolved ear recognition problems as well as ear databases available for researchers.

DeepFace: Closing the Gap to Human-Level Performance in Face Verification In modern face recognition, the conventional pipeline consists of four stages: detect => align => represent => classify. We revisit both the alignment step and the representation step by employing explicit 3D face modeling in order to apply a piecewise affine transformation, and derive a face representation from a nine-layer deep neural network. This deep network involves more than 120 million parameters using several locally connected layers without weight sharing, rather than the standard convolutional layers. Thus we trained it on the largest facial dataset to-date, an identity labeled dataset of four million facial images belonging to more than 4, 000 identities. The learned representations coupling the accurate model-based alignment with the large facial database generalize remarkably well to faces in unconstrained environments, even with a simple classifier. Our method reaches an accuracy of 97.35% on the Labeled Faces in the Wild (LFW) dataset, reducing the error of the current state of the art by more than 27%, closely approaching human-level performance.

Markov games as a framework for multi-agent reinforcement learning In the Markov decision process (MDP) formalization of reinforcement learning, a single adaptive agent interacts with an environment defined by a probabilistic transition function. In this solipsis-tic view, secondary agents can only be part of the environment and are therefore fixed in their behavior. The framework of Markov games allows us to widen this view to include multiple adaptive agents with interacting or competing goals. This paper considers a step in this direction in which exactly two agents with diametrically opposed goals share an environment. It describes a Q-learning-like algorithm for finding optimal policies and demonstrates its application to a simple two-player game in which the optimal policy is probabilistic.

Pors: proofs of retrievability for large files In this paper, we define and explore proofs of retrievability (PORs). A POR scheme enables an archive or back-up service (prover) to produce a concise proof that a user (verifier) can retrieve a target file F, that is, that the archive retains and reliably transmits file data sufficient for the user to recover F in its entirety. A POR may be viewed as a kind of cryptographic proof of knowledge (POK), but one specially designed to handle a large file (or bitstring) F. We explore POR protocols here in which the communication costs, number of memory accesses for the prover, and storage requirements of the user (verifier) are small parameters essentially independent of the length of F. In addition to proposing new, practical POR constructions, we explore implementation considerations and optimizations that bear on previously explored, related schemes. In a POR, unlike a POK, neither the prover nor the verifier need actually have knowledge of F. PORs give rise to a new and unusual security definition whose formulation is another contribution of our work. We view PORs as an important tool for semi-trusted online archives. Existing cryptographic techniques help users ensure the privacy and integrity of files they retrieve. It is also natural, however, for users to want to verify that archives do not delete or modify files prior to retrieval. The goal of a POR is to accomplish these checks without users having to download the files themselves. A POR can also provide quality-of-service guarantees, i.e., show that a file is retrievable within a certain time bound.

Design, Implementation, and Experimental Results of a Quaternion-Based Kalman Filter for Human Body Motion Tracking Real-time tracking of human body motion is an important technology in synthetic environments, robotics, and other human-computer interaction applications. This paper presents an extended Kalman filter designed for real-time estimation of the orientation of human limb segments. The filter processes data from small inertial/magnetic sensor modules containing triaxial angular rate sensors, accelerometers, and magnetometers. The filter represents rotation using quaternions rather than Euler angles or axis/angle pairs. Preprocessing of the acceleration and magnetometer measurements using the Quest algorithm produces a computed quaternion input for the filter. This preprocessing reduces the dimension of the state vector and makes the measurement equations linear. Real-time implementation and testing results of the quaternion-based Kalman filter are presented. Experimental results validate the filter design, and show the feasibility of using inertial/magnetic sensor modules for real-time human body motion tracking

Wireless Networks with RF Energy Harvesting: A Contemporary Survey Radio frequency (RF) energy transfer and harvesting techniques have recently become alternative methods to power the next generation wireless networks. As this emerging technology enables proactive energy replenishment of wireless devices, it is advantageous in supporting applications with quality of service (QoS) requirements. In this paper, we present a comprehensive literature review on the research progresses in wireless networks with RF energy harvesting capability, referred to as RF energy harvesting networks (RF-EHNs). First, we present an overview of the RF-EHNs including system architecture, RF energy harvesting techniques and existing applications. Then, we present the background in circuit design as well as the state-of-the-art circuitry implementations, and review the communication protocols specially designed for RF-EHNs. We also explore various key design issues in the development of RFEHNs according to the network types, i.e., single-hop networks, multi-antenna networks, relay networks, and cognitive radio networks. Finally, we envision some open research directions.

Automated Detection of Obstructive Sleep Apnea Events from a Single-Lead Electrocardiogram Using a Convolutional Neural Network. In this study, we propose a method for the automated detection of obstructive sleep apnea (OSA) from a single-lead electrocardiogram (ECG) using a convolutional neural network (CNN). A CNN model was designed with six optimized convolution layers including activation, pooling, and dropout layers. One-dimensional (1D) convolution, rectified linear units (ReLU), and max pooling were applied to the convolution, activation, and pooling layers, respectively. For training and evaluation of the CNN model, a single-lead ECG dataset was collected from 82 subjects with OSA and was divided into training (including data from 63 patients with 34,281 events) and testing (including data from 19 patients with 8571 events) datasets. Using this CNN model, a precision of 0.99%, a recall of 0.99%, and an F-score of 0.99% were attained with the training dataset; these values were all 0.96% when the CNN was applied to the testing dataset. These results show that the proposed CNN model can be used to detect OSA accurately on the basis of a single-lead ECG. Ultimately, this CNN model may be used as a screening tool for those suspected to suffer from OSA.

A Hierarchical Architecture Using Biased Min-Consensus for USV Path Planning This paper proposes a hierarchical architecture using the biased min-consensus (BMC) method, to solve the path planning problem of unmanned surface vessel (USV). We take the fixed-point monitoring mission as an example, where a series of intermediate monitoring points should be visited once by USV. The whole framework incorporates the low-level layer planning the standard path between any two intermediate points, and the high-level fashion determining their visiting sequence. First, the optimal standard path in terms of voyage time and risk measure is planned by the BMC protocol, given that the corresponding graph is constructed with node state and edge weight. The USV will avoid obstacles or keep a certain distance safely, and arrive at the target point quickly. It is proven theoretically that the state of the graph will converge to be stable after finite iterations, i.e., the optimal solution can be found by BMC with low calculation complexity. Second, by incorporating the constraint of intermediate points, their visiting sequence is optimized by BMC again with the reconstruction of a new virtual graph based on the former planned results. The extensive simulation results in various scenarios also validate the feasibility and effectiveness of our method for autonomous navigation.

Multiple knowledge representation for big data artificial intelligence: framework, applications, and case studies In this paper, we present a multiple knowledge representation (MKR) framework and discuss its potential for developing big data artificial intelligence (AI) techniques with possible broader impacts across different AI areas. Typically, canonical knowledge representations and modern representations each emphasize a particular aspect of transforming inputs into symbolic encoding or vectors. For example, knowledge graphs focus on depicting semantic connections among concepts, whereas deep neural networks (DNNs) are more of a tool to perceive raw signal inputs. MKR is an advanced AI representation framework for more complete intelligent functions, such as raw signal perception, feature extraction and vectorization, knowledge symbolization, and logical reasoning. MKR has two benefits: (1) it makes the current AI techniques (dominated by deep learning) more explainable and generalizable, and (2) it expands current AI techniques by integrating MKR to facilitate the mutual benefits of the complementary capacity of each representation, e.g., raw signal perception and symbolic encoding. We expect that MKR research and its applications will drive the evolution of AI 2.0 and beyond.

Analysing user physiological responses for affective video summarisation. Video summarisation techniques aim to abstract the most significant content from a video stream. This is typically achieved by processing low-level image, audio and text features which are still quite disparate from the high-level semantics that end users identify with (the ‘semantic gap’). Physiological responses are potentially rich indicators of memorable or emotionally engaging video content for a given user. Consequently, we investigate whether they may serve as a suitable basis for a video summarisation technique by analysing a range of user physiological response measures, specifically electro-dermal response (EDR), respiration amplitude (RA), respiration rate (RR), blood volume pulse (BVP) and heart rate (HR), in response to a range of video content in a variety of genres including horror, comedy, drama, sci-fi and action. We present an analysis framework for processing the user responses to specific sub-segments within a video stream based on percent rank value normalisation. The application of the analysis framework reveals that users respond significantly to the most entertaining video sub-segments in a range of content domains. Specifically, horror content seems to elicit significant EDR, RA, RR and BVP responses, and comedy content elicits comparatively lower levels of EDR, but does seem to elicit significant RA, RR, BVP and HR responses. Drama content seems to elicit less significant physiological responses in general, and both sci-fi and action content seem to elicit significant EDR responses. We discuss the implications this may have for future affective video summarisation approaches.

On the roles of eye gaze and head dynamics in predicting driver's intent to change lanes Driver behavioral cues may present a rich source of information and feedback for future intelligent advanced driver-assistance systems (ADASs). With the design of a simple and robust ADAS in mind, we are interested in determining the most important driver cues for distinguishing driver intent. Eye gaze may provide a more accurate proxy than head movement for determining driver attention, whereas the measurement of head motion is less cumbersome and more reliable in harsh driving conditions. We use a lane-change intent-prediction system (McCall et al., 2007) to determine the relative usefulness of each cue for determining intent. Various combinations of input data are presented to a discriminative classifier, which is trained to output a prediction of probable lane-change maneuver at a particular point in the future. Quantitative results from a naturalistic driving study are presented and show that head motion, when combined with lane position and vehicle dynamics, is a reliable cue for lane-change intent prediction. The addition of eye gaze does not improve performance as much as simpler head dynamics cues. The advantage of head data over eye data is shown to be statistically significant (p

Detection of Driver Fatigue Caused by Sleep Deprivation This paper aims to provide reliable indications of driver drowsiness based on the characteristics of driver-vehicle interaction. A test bed was built under a simulated driving environment, and a total of 12 subjects participated in two experiment sessions requiring different levels of sleep (partial sleep-deprivation versus no sleep-deprivation) before the experiment. The performance of the subjects was analyzed in a series of stimulus-response and routine driving tasks, which revealed the performance differences of drivers under different sleep-deprivation levels. The experiments further demonstrated that sleep deprivation had greater effect on rule-based than on skill-based cognitive functions: when drivers were sleep-deprived, their performance of responding to unexpected disturbances degraded, while they were robust enough to continue the routine driving tasks such as lane tracking, vehicle following, and lane changing. In addition, we presented both qualitative and quantitative guidelines for designing drowsy-driver detection systems in a probabilistic framework based on the paradigm of Bayesian networks. Temporal aspects of drowsiness and individual differences of subjects were addressed in the framework.

Online Prediction of Driver Distraction Based on Brain Activity Patterns This paper presents a new computational framework for early detection of driver distractions (map viewing) using brain activity measured by electroencephalographic (EEG) signals. Compared with most studies in the literature, which are mainly focused on the classification of distracted and nondistracted periods, this study proposes a new framework to prospectively predict the start and end of a distraction period, defined by map viewing. The proposed prediction algorithm was tested on a data set of continuous EEG signals recorded from 24 subjects. During the EEG recordings, the subjects were asked to drive from an initial position to a destination using a city map in a simulated driving environment. The overall accuracy values for the prediction of the start and the end of map viewing were 81% and 70%, respectively. The experimental results demonstrated that the proposed algorithm can predict the start and end of map viewing with relatively high accuracy and can be generalized to individual subjects. The outcome of this study has a high potential to improve the design of future intelligent navigation systems. Prediction of the start of map viewing can be used to provide route information based on a driver's needs and consequently avoid map-viewing activities. Prediction of the end of map viewing can be used to provide warnings for potential long map-viewing durations. Further development of the proposed framework and its applications in driver-distraction predictions are also discussed.

Keep Your Scanners Peeled: Gaze Behavior as a Measure of Automation Trust During Highly Automated Driving. Objective: The feasibility of measuring drivers' automation trust via gaze behavior during highly automated driving was assessed with eye tracking and validated with self-reported automation trust in a driving simulator study. Background: Earlier research from other domains indicates that drivers' automation trust might be inferred from gaze behavior, such as monitoring frequency. Method: The gaze behavior and self-reported automation trust of 35 participants attending to a visually demanding non-driving-related task (NDRT) during highly automated driving was evaluated. The relationship between dispositional, situational, and learned automation trust with gaze behavior was compared. Results: Overall, there was a consistent relationship between drivers' automation trust and gaze behavior. Participants reporting higher automation trust tended to monitor the automation less frequently. Further analyses revealed that higher automation trust was associated with lower monitoring frequency of the automation during NDRTs, and an increase in trust over the experimental session was connected with a decrease in monitoring frequency. Conclusion: We suggest that (a) the current results indicate a negative relationship between drivers' self-reported automation trust and monitoring frequency, (b) gaze behavior provides a more direct measure of automation trust than other behavioral measures, and (c) with further refinement, drivers' automation trust during highly automated driving might be inferred from gaze behavior. Application: Potential applications of this research include the estimation of drivers' automation trust and reliance during highly automated driving.

3D separable convolutional neural network for dynamic hand gesture recognition. •The Frame Difference method is used to pre-process the input in order to filter the background.•A 3D separable CNN is proposed for dynamic gesture recognition. The standard 3D convolution process is decomposed into two processes: 3D depth-wise and 3D point-wise.•By the application of skip connection and layer-wise learning rate, the undesirable gradient dispersion due to the separation operation is solved and the performance of the network is improved.•A dynamic hand gesture library is built through HoloLens.

Deep convolutional neural network-based Bernoulli heatmap for head pose estimation Head pose estimation is a crucial problem for many tasks, such as driver attention, fatigue detection, and human behaviour analysis. It is well known that neural networks are better at handling classification problems than regression problems. It is an extremely nonlinear process to let the network output the angle value directly for optimization learning, and the weight constraint of the loss function will be relatively weak. This paper proposes a novel Bernoulli heatmap for head pose estimation from a single RGB image. Our method can achieve the positioning of the head area while estimating the angles of the head. The Bernoulli heatmap makes it possible to construct fully convolutional neural networks without fully connected layers and provides a new idea for the output form of head pose estimation. A deep convolutional neural network (CNN) structure with multiscale representations is adopted to maintain high-resolution information and low-resolution information in parallel. This kind of structure can maintain rich, high-resolution representations. In addition, channelwise fusion is adopted to make the fusion weights learnable instead of simple addition with equal weights. As a result, the estimation is spatially more precise and potentially more accurate. The effectiveness of the proposed method is empirically demonstrated by comparing it with other state-of-the-art methods on public datasets.

Reinforcement learning based data fusion method for multi-sensors In order to improve detection system robustness and reliability, multi-sensors fusion is used in modern air combat. In this paper, a data fusion method based on reinforcement learning is developed for multi-sensors. Initially, the cubic B-spline interpolation is used to solve time alignment problems of multisource data. Then, the reinforcement learning based data fusion (RLBDF) method is proposed to obtain the fusion results. With the case that the priori knowledge of target is obtained, the fusion accuracy reinforcement is realized by the error between fused value and actual value. Furthermore, the Fisher information is instead used as the reward if the priori knowledge is unable to be obtained. Simulations results verify that the developed method is feasible and effective for the multi-sensors data fusion in air combat.

Mobile Edge Computing Enabled 5G Health Monitoring for Internet of Medical Things: A Decentralized Game Theoretic Approach The prompt evolution of Internet of Medical Things (IoMT) promotes pervasive in-home health monitoring networks. However, excessive requirements of patients result in insufficient spectrum resources and communication overload. Mobile Edge Computing (MEC) enabled 5G health monitoring is conceived as a favorable paradigm to tackle such an obstacle. In this paper, we construct a cost-efficient in-home health monitoring system for IoMT by dividing it into two sub-networks, i.e., intra-Wireless Body Area Networks (WBANs) and beyond-WBANs. Highlighting the characteristics of IoMT, the cost of patients depends on medical criticality, Age of Information (AoI) and energy consumption. For intra-WBANs, a cooperative game is formulated to allocate the wireless channel resources. While for beyond-WBANs, considering the individual rationality and potential selfishness, a decentralized non-cooperative game is proposed to minimize the system-wide cost in IoMT. We prove that the proposed algorithm can reach a Nash equilibrium. In addition, the upper bound of the algorithm time complexity and the number of patients benefiting from MEC is theoretically derived. Performance evaluations demonstrate the effectiveness of our proposed algorithm with respect to the system-wide cost and the number of patients benefiting from MEC.

Artificial fish swarm algorithm: a survey of the state-of-the-art, hybridization, combinatorial and indicative applications FSA (artificial fish-swarm algorithm) is one of the best methods of optimization among the swarm intelligence algorithms. This algorithm is inspired by the collective movement of the fish and their various social behaviors. Based on a series of instinctive behaviors, the fish always try to maintain their colonies and accordingly demonstrate intelligent behaviors. Searching for food, immigration and dealing with dangers all happen in a social form and interactions between all fish in a group will result in an intelligent social behavior.This algorithm has many advantages including high convergence speed, flexibility, fault tolerance and high accuracy. This paper is a review of AFSA algorithm and describes the evolution of this algorithm along with all improvements, its combination with various methods as well as its applications. There are many optimization methods which have a affinity with this method and the result of this combination will improve the performance of this method. Its disadvantages include high time complexity, lack of balance between global and local search, in addition to lack of benefiting from the experiences of group members for the next movements.

Short-Term Traffic Flow Forecasting: An Experimental Comparison of Time-Series Analysis and Supervised Learning The literature on short-term traffic flow forecasting has undergone great development recently. Many works, describing a wide variety of different approaches, which very often share similar features and ideas, have been published. However, publications presenting new prediction algorithms usually employ different settings, data sets, and performance measurements, making it difficult to infer a clear picture of the advantages and limitations of each model. The aim of this paper is twofold. First, we review existing approaches to short-term traffic flow forecasting methods under the common view of probabilistic graphical models, presenting an extensive experimental comparison, which proposes a common baseline for their performance analysis and provides the infrastructure to operate on a publicly available data set. Second, we present two new support vector regression models, which are specifically devised to benefit from typical traffic flow seasonality and are shown to represent an interesting compromise between prediction accuracy and computational efficiency. The SARIMA model coupled with a Kalman filter is the most accurate model; however, the proposed seasonal support vector regressor turns out to be highly competitive when performing forecasts during the most congested periods.

TSCA: A Temporal-Spatial Real-Time Charging Scheduling Algorithm for On-Demand Architecture in Wireless Rechargeable Sensor Networks. The collaborative charging issue in Wireless Rechargeable Sensor Networks (WRSNs) is a popular research problem. With the help of wireless power transfer technology, electrical energy can be transferred from wireless charging vehicles (WCVs) to sensors, providing a new paradigm to prolong network lifetime. Existing techniques on collaborative charging usually take the periodical and deterministic approach, but neglect influences of non-deterministic factors such as topological changes and node failures, making them unsuitable for large-scale WRSNs. In this paper, we develop a temporal-spatial charging scheduling algorithm, namely TSCA, for the on-demand charging architecture. We aim to minimize the number of dead nodes while maximizing energy efficiency to prolong network lifetime. First, after gathering charging requests, a WCV will compute a feasible movement solution. A basic path planning algorithm is then introduced to adjust the charging order for better efficiency. Furthermore, optimizations are made in a global level. Then, a node deletion algorithm is developed to remove low efficient charging nodes. Lastly, a node insertion algorithm is executed to avoid the death of abandoned nodes. Extensive simulations show that, compared with state-of-the-art charging scheduling algorithms, our scheme can achieve promising performance in charging throughput, charging efficiency, and other performance metrics.

A novel adaptive dynamic programming based on tracking error for nonlinear discrete-time systems In this paper, to eliminate the tracking error by using adaptive dynamic programming (ADP) algorithms, a novel formulation of the value function is presented for the optimal tracking problem (TP) of nonlinear discrete-time systems. Unlike existing ADP methods, this formulation introduces the control input into the tracking error, and ignores the quadratic form of the control input directly, which makes the boundedness and convergence of the value function independent of the discount factor. Based on the proposed value function, the optimal control policy can be deduced without considering the reference control input. Value iteration (VI) and policy iteration (PI) methods are applied to prove the optimality of the obtained control policy, and derived the monotonicity property and convergence of the iterative value function. Simulation examples realized with neural networks and the actor–critic structure are provided to verify the effectiveness of the proposed ADP algorithm.

Intelligent edge computing based on machine learning for smart city To alleviate the huge computing pressure caused by the single mobile edge server computing mode as the amount of data increases, in this research, we propose a method to conduct calculations in a collaborative way. First, the method needs to consider how to encourage devices to cooperate when they are selfish. Second, the method answers the following question: how can collaborative computing be carried out when the device has the intention to cooperate? For example, how can calculations be conducted when there are extensibility and privacy problems in machine learning tasks? In view of the above challenges, a mobile edge server is taken as the focus, and the available resources around the mobile edge server are used for collaborative computing to further improve the computing performance of a mobile edge computing (MEC) system. The alternating direction multiplier method is used to solve the problem. First, the relevant techniques and theories of MEC, Stackelberg principle-subordinate game theory, and the alternating direction method of multipliers (ADMM) are introduced. Then, the problem description and model construction of distributed task scheduling in MEC and machine learning task-based device coordination computing are introduced, and machine learning is applied in the distributed task scheduling algorithm and distributed device coordination algorithm. Finally, the distributed task scheduling algorithm and distributed device coordination algorithm are tested by experiments.

Delay-Aware Microservice Coordination in Mobile Edge Computing: A Reinforcement Learning Approach As an emerging service architecture, microservice enables decomposition of a monolithic web service into a set of independent lightweight services which can be executed independently. With mobile edge computing, microservices can be further deployed in edge clouds dynamically, launched quickly, and migrated across edge clouds easily, providing better services for users in proximity. However, the user mobility can result in frequent switch of nearby edge clouds, which increases the service delay when users move away from their serving edge clouds. To address this issue, this article investigates microservice coordination among edge clouds to enable seamless and real-time responses to service requests from mobile users. The objective of this work is to devise the optimal microservice coordination scheme which can reduce the overall service delay with low costs. To this end, we first propose a dynamic programming-based offline microservice coordination algorithm, that can achieve the globally optimal performance. However, the offline algorithm heavily relies on the availability of the prior information such as computation request arrivals, time-varying channel conditions and edge cloud's computation capabilities required, which is hard to be obtained. Therefore, we reformulate the microservice coordination problem using Markov decision process framework and then propose a reinforcement learning-based online microservice coordination algorithm to learn the optimal strategy. Theoretical analysis proves that the offline algorithm can find the optimal solution while the online algorithm can achieve near-optimal performance. Furthermore, based on two real-world datasets, i.e., the Telecom's base station dataset and Taxi Track dataset from Shanghai, experiments are conducted. The experimental results demonstrate that the proposed online algorithm outperforms existing algorithms in terms of service delay and migration costs, and the achieved performance is close to the optimal performance obtained by the offline algorithm.

Energy-Aware Task Offloading and Resource Allocation for Time-Sensitive Services in Mobile Edge Computing Systems Mobile Edge Computing (MEC) is a promising architecture to reduce the energy consumption of mobile devices and provide satisfactory quality-of-service to time-sensitive services. How to jointly optimize task offloading and resource allocation to minimize the energy consumption subject to the latency requirement remains an open problem, which motivates this paper. When the latency constraint is tak...

Dynamic Deployment and Cost-Sensitive Provisioning for Elastic Mobile Cloud Services. As mobile customers gradually occupying the largest share of cloud service users, the effective and cost-sensitive provisioning of mobile cloud services quickly becomes a main theme in cloud computing. The key issues involved are much more than just enabling mobile users to access remote cloud resources through wireless networks. The resource limited and intermittent disconnection problems of mobile environments have intrinsic conflict with the continuous connection assumption of the cloud service usage patterns. We advocate that seamless service provisioning in mobile cloud can only be achieved with full exploitation of all available resources around mobile users. An elastic framework is proposed to automatically and dynamically deploy cloud services on data center, base stations, client units, even peer devices. The best deployment location is dynamically determined based on a context-aware and cost-sensitive evaluation model. To facilitate easy adoption of the proposed framework, a service development model and associated semi-automatic tools are provided such that cloud service developers can easily convert a service for execution on different platforms without porting. Prototype implementation and evaluation on the Google Cloud and Android platforms demonstrate that our mechanism can successfully maintain seamless services with very low overhead.

Edge-Computing-Enabled Smart Cities: A Comprehensive Survey Recent years have disclosed a remarkable proliferation of compute-intensive applications in smart cities. Such applications continuously generate enormous amounts of data which demand strict latency-aware computational processing capabilities. Although edge computing is an appealing technology to compensate for stringent latency-related issues, its deployment engenders new challenges. In this article, we highlight the role of edge computing in realizing the vision of smart cities. First, we analyze the evolution of edge computing paradigms. Subsequently, we critically review the state-of-the-art literature focusing on edge computing applications in smart cities. Later, we categorize and classify the literature by devising a comprehensive and meticulous taxonomy. Furthermore, we identify and discuss key requirements, and enumerate recently reported synergies of edge computing-enabled smart cities. Finally, several indispensable open challenges along with their causes and guidelines are discussed, serving as future research directions.

A Cooperative Resource Allocation Model For Iot Applications In Mobile Edge Computing With the advancement in the development of the Internet of Things (IoT) technology, as well as the industrial IoT, various applications and services are benefiting from this emerging technology such as smart healthcare systems, virtual realities applications, connected and autonomous vehicles, to name a few. However, IoT devices are known for being limited computation capacities which is crucial to the device's availability time. Traditional approaches used to offload the applications to the cloud to ease the burden on the end user's devices, however, greater latency and network traffic issues still persist. Mobile Edge Computing (MEC) technology has emerged to address these issues and enhance the survivability of cloud infrastructure. While a lot of attempts have been made to manage an efficient process of applications offload, many of which either focus on the allocation of computational or communication protocols without considering a cooperative solution. In addition, a single-user scenario was considered. Therefore, we study multi-user IoT applications offloading for a MEC system, which cooperatively considers to allocate both the resources of computation and communication. The proposed system focuses on minimizing the weighted overhead of local IoT devices, and minimize the offload measured by the delay and energy consumption. The mathematical formulation is a typical mixed integer nonlinear programming (MINP), and this is an NP-hard problem. We obtain the solution to the objective function by splitting the objective problem into three sub-problems. Extensive set of evaluations have been performed so as to get the evaluation of the proposed model. The collected results indicate that offloading decisions, energy consumption, latency, and the impact of the number of IoT devices have shown superior improvement over traditional models.

Probabilistic encryption A new probabilistic model of data encryption is introduced. For this model, under suitable complexity assumptions, it is proved that extracting any information about the cleartext from the cyphertext is hard on the average for an adversary with polynomially bounded computational resources. The proof holds for any message space with any probability distribution. The first implementation of this model is presented. The security of this implementation is proved under the interactability assumptin of deciding Quadratic Residuosity modulo composite numbers whose factorization is unknown.

A powerful and efficient algorithm for numerical function optimization: artificial bee colony (ABC) algorithm Swarm intelligence is a research branch that models the population of interacting agents or swarms that are able to self-organize. An ant colony, a flock of birds or an immune system is a typical example of a swarm system. Bees' swarming around their hive is another example of swarm intelligence. Artificial Bee Colony (ABC) Algorithm is an optimization algorithm based on the intelligent behaviour of honey bee swarm. In this work, ABC algorithm is used for optimizing multivariable functions and the results produced by ABC, Genetic Algorithm (GA), Particle Swarm Algorithm (PSO) and Particle Swarm Inspired Evolutionary Algorithm (PS-EA) have been compared. The results showed that ABC outperforms the other algorithms.

Markov games as a framework for multi-agent reinforcement learning In the Markov decision process (MDP) formalization of reinforcement learning, a single adaptive agent interacts with an environment defined by a probabilistic transition function. In this solipsis-tic view, secondary agents can only be part of the environment and are therefore fixed in their behavior. The framework of Markov games allows us to widen this view to include multiple adaptive agents with interacting or competing goals. This paper considers a step in this direction in which exactly two agents with diametrically opposed goals share an environment. It describes a Q-learning-like algorithm for finding optimal policies and demonstrates its application to a simple two-player game in which the optimal policy is probabilistic.

Scalable and efficient provable data possession. Storage outsourcing is a rising trend which prompts a number of interesting security issues, many of which have been extensively investigated in the past. However, Provable Data Possession (PDP) is a topic that has only recently appeared in the research literature. The main issue is how to frequently, efficiently and securely verify that a storage server is faithfully storing its client's (potentially very large) outsourced data. The storage server is assumed to be untrusted in terms of both security and reliability. (In other words, it might maliciously or accidentally erase hosted data; it might also relegate it to slow or off-line storage.) The problem is exacerbated by the client being a small computing device with limited resources. Prior work has addressed this problem using either public key cryptography or requiring the client to outsource its data in encrypted form. In this paper, we construct a highly efficient and provably secure PDP technique based entirely on symmetric key cryptography, while not requiring any bulk encryption. Also, in contrast with its predecessors, our PDP technique allows outsourcing of dynamic data, i.e, it efficiently supports operations, such as block modification, deletion and append.

Cognitive Cars: A New Frontier for ADAS Research This paper provides a survey of recent works on cognitive cars with a focus on driver-oriented intelligent vehicle motion control. The main objective here is to clarify the goals and guidelines for future development in the area of advanced driver-assistance systems (ADASs). Two major research directions are investigated and discussed in detail: 1) stimuli–decisions–actions, which focuses on the driver side, and 2) perception enhancement–action-suggestion–function-delegation, which emphasizes the ADAS side. This paper addresses the important achievements and major difficulties of each direction and discusses how to combine the two directions into a single integrated system to obtain safety and comfort while driving. Other related topics, including driver training and infrastructure design, are also studied.

Online Prediction of Driver Distraction Based on Brain Activity Patterns This paper presents a new computational framework for early detection of driver distractions (map viewing) using brain activity measured by electroencephalographic (EEG) signals. Compared with most studies in the literature, which are mainly focused on the classification of distracted and nondistracted periods, this study proposes a new framework to prospectively predict the start and end of a distraction period, defined by map viewing. The proposed prediction algorithm was tested on a data set of continuous EEG signals recorded from 24 subjects. During the EEG recordings, the subjects were asked to drive from an initial position to a destination using a city map in a simulated driving environment. The overall accuracy values for the prediction of the start and the end of map viewing were 81% and 70%, respectively. The experimental results demonstrated that the proposed algorithm can predict the start and end of map viewing with relatively high accuracy and can be generalized to individual subjects. The outcome of this study has a high potential to improve the design of future intelligent navigation systems. Prediction of the start of map viewing can be used to provide route information based on a driver's needs and consequently avoid map-viewing activities. Prediction of the end of map viewing can be used to provide warnings for potential long map-viewing durations. Further development of the proposed framework and its applications in driver-distraction predictions are also discussed.

Adaptive fuzzy tracking control for switched uncertain strict-feedback nonlinear systems. •Adaptive tracking control for switched strict-feedback nonlinear systems is proposed.•The generalized fuzzy hyperbolic model is used to approximate nonlinear functions.•The designed controller has fewer design parameters comparing with existing methods.

Social Robots for (Second) Language Learning in (Migrant) Primary School Children Especially these days, innovation and support from technology to relieve pressure in education is highly urgent. This study tested the potential advantage of a social robot over a tablet in (second) language learning on performance, engagement, and enjoyment. Shortages in primary education call for new technology solutions. Previous studies combined robots with tablets, to compensate for robot’s limitations, however, this study applied direct human–robot interaction. Primary school children (N = 63, aged 4–6) participated in a 3-wave field experiment with story-telling exercises, either with a semi-autonomous robot (without tablet, using WOz) or a tablet. Results showed increased learning gains over time when training with a social robot, compared to the tablet. Children who trained with a robot were more engaged in the story-telling task and enjoyed it more. Robot’s behavioral style (social or neutral) hardly differed overall, however, seems to vary for high versus low educational abilities. While social robots need sophistication before being implemented in schools, our study shows the potential of social robots as tutors in (second) language learning.

Self-aligning exoskeleton axes through decoupling of joint rotations and translations To automatically align exoskeleton axes to human anatomical axes, we propose to decouple the joint rotations from the joint translations. Decoupling can reduce setup times and painful misalignment forces, at the cost of increased mechanical complexity and movement inertia. The decoupling approach was applied to the Dampace and Limpact exoskeletons.

Wearable soft sensing suit for human gait measurement Wearable robots based on soft materials will augment mobility and performance of the host without restricting natural kinematics. Such wearable robots will need soft sensors to monitor the movement of the wearer and robot outside the lab. Until now wearable soft sensors have not demonstrated significant mechanical robustness nor been systematically characterized for human motion studies of walking and running. Here, we present the design and systematic characterization of a soft sensing suit for monitoring hip, knee, and ankle sagittal plane joint angles. We used hyper-elastic strain sensors based on microchannels of liquid metal embedded within elastomer, but refined their design with the use of discretized stiffness gradients to improve mechanical durability. We found that these robust sensors could stretch up to 396% of their original lengths, would restrict the wearer by less than 0.17% of any given joint's torque, had gauge factor sensitivities of greater than 2.2, and exhibited less than 2% change in electromechanical specifications through 1500 cycles of loading-unloading. We also evaluated the accuracy and variability of the soft sensing suit by comparing it with joint angle data obtained through optical motion capture. The sensing suit had root mean square (RMS) errors of less than 5° for a walking speed of 0.89 m/s and reached a maximum RMS error of 15° for a running speed of 2.7 m/s. Despite the deviation of absolute measure, the relative repeatability of the sensing suit's joint angle measurements were statistically equivalent to that of optical motion capture at all speeds. We anticipate that wearable soft sensing will also have applications beyond wearable robotics, such as in medical diagnostics and in human-computer interaction.

Effects of robotic knee exoskeleton on human energy expenditure. A number of studies discuss the design and control of various exoskeleton mechanisms, yet relatively few address the effect on the energy expenditure of the user. In this paper, we discuss the effect of a performance augmenting exoskeleton on the metabolic cost of an able-bodied user/pilot during periodic squatting. We investigated whether an exoskeleton device will significantly reduce the metabolic cost and what is the influence of the chosen device control strategy. By measuring oxygen consumption, minute ventilation, heart rate, blood oxygenation, and muscle EMG during 5-min squatting series, at one squat every 2 s, we show the effects of using a prototype robotic knee exoskeleton under three different noninvasive control approaches: gravity compensation approach, position-based approach, and a novel oscillator-based approach. The latter proposes a novel control that ensures synchronization of the device and the user. Statistically significant decrease in physiological responses can be observed when using the robotic knee exoskeleton under gravity compensation and oscillator-based control. On the other hand, the effects of position-based control were not significant in all parameters although all approaches significantly reduced the energy expenditure during squatting.

Multi-joint actuation platform for lower extremity soft exosuits Lower-limb wearable robots have been proposed as a means to augment or assist the wearer's natural performance, in particular, in the military and medical field. Previous research studies on human-robot interaction and biomechanics have largely been performed with rigid exoskeletons that add significant inertia to the lower extremities and provide constraints to the wearer's natural kinematics in both actuated and non-actuated degrees of freedom. Actuated lightweight soft exosuits minimize these effects and provide a unique opportunity to study human-robot interaction in wearable systems without affecting the subjects underlying natural dynamics. In this paper, we present the design and control of a reconfigurable multi-joint actuation platform that can provide biologically realistic torques to ankle, knee, and hip joints through lower extremity soft exosuits. Two different soft exosuits have been designed to deliver assistive forces through Bowden cable transmission to the ankle and hip joints. Through human subject experiments, it is demonstrated that with a real-time admittance controller, accurate force profile tracking can be achieved during walking. The average energy delivered to the test subject was calculated while walking at 1.25 m/s and actuated with 15% of the total torque required by the biological joints. The results show that the ankle joint received an average of 3.02J during plantar flexion and that the hip joint received 1.67J during flexion each gait cycle. The efficiency of the described suit and controller in transferring energy to the human biological joints is 70% for the ankle and 48% for the hip.

Biologically-inspired soft exosuit. In this paper, we present the design and evaluation of a novel soft cable-driven exosuit that can apply forces to the body to assist walking. Unlike traditional exoskeletons which contain rigid framing elements, the soft exosuit is worn like clothing, yet can generate moments at the ankle and hip with magnitudes of 18% and 30% of those naturally generated by the body during walking, respectively. Our design uses geared motors to pull on Bowden cables connected to the suit near the ankle. The suit has the advantages over a traditional exoskeleton in that the wearer's joints are unconstrained by external rigid structures, and the worn part of the suit is extremely light, which minimizes the suit's unintentional interference with the body's natural biomechanics. However, a soft suit presents challenges related to actuation force transfer and control, since the body is compliant and cannot support large pressures comfortably. We discuss the design of the suit and actuation system, including principles by which soft suits can transfer force to the body effectively and the biological inspiration for the design. For a soft exosuit, an important design parameter is the combined effective stiffness of the suit and its interface to the wearer. We characterize the exosuit's effective stiffness, and present preliminary results from it generating assistive torques to a subject during walking. We envision such an exosuit having broad applicability for assisting healthy individuals as well as those with muscle weakness.

Design of a Bio-Inspired Wearable Exoskeleton for Applications in Robotics In this paper we explain the methodology we adopted to design the kinematics structure of a multi-contact points haptic interface. We based our concept on the analysis of the human arm anatomy and kinematics with the intend to synthesize a system that will be able to interface with the human limb in a very natural way. We proposed a simplified kinematic model of the human arm using a notation coming from the robotics field. To find out the best kinematics architecture we employed real movement data, measured from a human subject, and integrated them with the kinematic model of the exoskeleton, this allow us to test the system before its construction and to formalize specific requirements. We also implemented and tested a first passive version of the shoulder joint.

Human Shoulder Modeling Including Scapulo-Thoracic Constraint And Joint Sinus Cones In virtual human modeling, the shoulder is usually composed of clavicular, scapular and arm segments related by rotational joints. Although the model is improved, the realistic animation of the shoulder is hardly achieved. This is due to the fact that it is difficult to coordinate the simultaneous motion of the shoulder components in a consistent way. Also, the common use of independent one-degree of freedom (DOF) joint hierarchies does not properly render the 3-D accessibility space of real joints. On the basis of former biomechanical investigations, we propose here an extended shoulder model including scapulo-thoracic constraint and joint sinus cones. As a demonstration, the model is applied, using inverse kinematics, to the animation of a 3-D anatomic muscled skeleton model. (C) 2000 Elsevier Science Ltd. All rights reserved.

Fabrication of “18 Weave” Muscles and Their Application to Soft Power Support Suit for Upper Limbs Using Thin McKibben Muscle In this letter, we propose a novel active weave named “18 Weave” using a thin McKibben muscle and report on the application to a prototype soft power support suit for the upper limb. The active weave utilizes the flexibility of the thin McKibben muscle, which is used for warps and wefts. From the proposed structure, we confirmed that a 19.4% additional displacement is achieved compared to a single...

Constrained Kalman filtering for indoor localization of transport vehicles using floor-installed HF RFID transponders Localization of transport vehicles is an important issue for many intralogistics applications. The paper presents an inexpensive solution for indoor localization of vehicles. Global localization is realized by detection of RFID transponders, which are integrated in the floor. The paper presents a novel algorithm for fusing RFID readings with odometry using Constraint Kalman filtering. The paper presents experimental results with a Mecanum based omnidirectional vehicle on a NaviFloor® installation, which includes passive HF RFID transponders. The experiments show that the proposed Constraint Kalman filter provides a similar localization accuracy compared to a Particle filter but with much lower computational expense.

A robust image authentication method distinguishing JPEG compression from malicious manipulation Image authentication verifies the originality of an image by detecting malicious manipulations. Its goal is different from that of image watermarking, which embeds into the image a signature surviving most manipulations. Most existing methods for image authentication treat all types of manipulation equally (i.e., as unacceptable). However, some practical applications demand techniques that can distinguish acceptable manipulations (e.g., compression) from malicious ones. In this paper, we present an effective technique for image authentication which can prevent malicious manipulations but allow JPEG lossy compression. The authentication signature is based on the invariance of the relationships between discrete cosine transform (DCT) coefficients at the same position in separate blocks of an image. These relationships are preserved when DCT coefficients are quantized in JPEG compression. Our proposed method can distinguish malicious manipulations from JPEG lossy compression regardless of the compression ratio or the number of compression iterations. We describe adaptive methods with probabilistic guarantee to handle distortions introduced by various acceptable manipulations such as integer rounding, image filtering, image enhancement, or scaling-recaling. We also present theoretical and experimental results to demonstrate the effectiveness of the technique

Tabu search based multi-watermarks embedding algorithm with multiple description coding Digital watermarking is a useful solution for digital rights management systems, and it has been a popular research topic in the last decade. Most watermarking related literature focuses on how to resist deliberate attacks by applying benchmarks to watermarked media that assess the effectiveness of the watermarking algorithm. Only a few papers have concentrated on the error-resilient transmission of watermarked media. In this paper, we propose an innovative algorithm for vector quantization (VQ) based image watermarking, which is suitable for error-resilient transmission over noisy channels. By incorporating watermarking with multiple description coding (MDC), the scheme we propose to embed multiple watermarks can effectively overcome channel impairments while retaining the capability for copyright and ownership protection. In addition, we employ an optimization technique, called tabu search, to optimize both the watermarked image quality and the robustness of the extracted watermarks. We have obtained promising simulation results that demonstrate the utility and practicality of our algorithm. (C) 2011 Elsevier Inc. All rights reserved.

Micro aerial vehicle networks: an experimental analysis of challenges and opportunities The need for aerial networks is growing with the recent advance of micro aerial vehicles, which enable a wide range of civilian applications. Our experimental analysis shows that wireless connectivity among MAVs is challenged by the mobility and heterogeneity of the nodes, lightweight antenna design, body blockage, constrained embedded resources, and limited battery power. However, the movement and location of MAVs are known and may be controlled to establish wireless links with the best transmission opportunities in time and space. This special ecosystem undoubtedly requires a rethinking of wireless communications and calls for novel networking approaches. Supported by empirical results, we identify important research questions, and introduce potential solutions and directions for investigation.

Adaptive fuzzy tracking control for switched uncertain strict-feedback nonlinear systems. •Adaptive tracking control for switched strict-feedback nonlinear systems is proposed.•The generalized fuzzy hyperbolic model is used to approximate nonlinear functions.•The designed controller has fewer design parameters comparing with existing methods.

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

Efficient energy management in wireless rechargeable sensor networks Through recent technology advances in the field of wireless energy transmission, Wireless Rechargeable Sensor Networks (WRSN) have emerged. In this new paradigm for WSNs a mobile entity called Mobile Charger (MC) traverses the network and replenishes the dissipated energy of sensors. In this work we first provide a formal definition of the charging dispatch decision problem and prove its computational hardness. We then investigate how to optimize the trade-offs of several critical aspects of the charging process such as a) the trajectory of the charger, b) the different charging policies and c) the impact of the ratio of the energy the MC may deliver to the sensors over the total available energy in the network. In the light of these optimizations, we then study the impact of the charging process to the network lifetime for three characteristic underlying routing protocols; a greedy protocol, a clustering protocol and an energy balancing protocol. Finally, we propose a Mobile Charging Protocol that locally adapts the circular trajectory of the MC to the energy dissipation rate of each sub-region of the network. We compare this protocol against several MC trajectories for all three routing families by a detailed experimental evaluation. The derived findings demonstrate significant performance gains, both with respect to the no charger case as well as the different charging alternatives; in particular, the performance improvements include the network lifetime, as well as connectivity, coverage and energy balance properties.

Mobility in wireless sensor networks - Survey and proposal. Targeting an increasing number of potential application domains, wireless sensor networks (WSN) have been the subject of intense research, in an attempt to optimize their performance while guaranteeing reliability in highly demanding scenarios. However, hardware constraints have limited their application, and real deployments have demonstrated that WSNs have difficulties in coping with complex communication tasks – such as mobility – in addition to application-related tasks. Mobility support in WSNs is crucial for a very high percentage of application scenarios and, most notably, for the Internet of Things. It is, thus, important to know the existing solutions for mobility in WSNs, identifying their main characteristics and limitations. With this in mind, we firstly present a survey of models for mobility support in WSNs. We then present the Network of Proxies (NoP) assisted mobility proposal, which relieves resource-constrained WSN nodes from the heavy procedures inherent to mobility management. The presented proposal was implemented and evaluated in a real platform, demonstrating not only its advantages over conventional solutions, but also its very good performance in the simultaneous handling of several mobile nodes, leading to high handoff success rate and low handoff time.

Tag-based cooperative data gathering and energy recharging in wide area RFID sensor networks The Wireless Identification and Sensing Platform (WISP) conjugates the identification potential of the RFID technology and the sensing and computing capability of the wireless sensors. Practical issues, such as the need of periodically recharging WISPs, challenge the effective deployment of large-scale RFID sensor networks (RSNs) consisting of RFID readers and WISP nodes. In this view, the paper proposes cooperative solutions to energize the WISP devices in a wide-area sensing network while reducing the data collection delay. The main novelty is the fact that both data transmissions and energy transfer are based on the RFID technology only: RFID mobile readers gather data from the WISP devices, wirelessly recharge them, and mutually cooperate to reduce the data delivery delay to the sink. Communication between mobile readers relies on two proposed solutions: a tag-based relay scheme, where RFID tags are exploited to temporarily store sensed data at pre-determined contact points between the readers; and a tag-based data channel scheme, where the WISPs are used as a virtual communication channel for real time data transfer between the readers. Both solutions require: (i) clustering the WISP nodes; (ii) dimensioning the number of required RFID mobile readers; (iii) planning the tour of the readers under the energy and time constraints of the nodes. A simulative analysis demonstrates the effectiveness of the proposed solutions when compared to non-cooperative approaches. Differently from classic schemes in the literature, the solutions proposed in this paper better cope with scalability issues, which is of utmost importance for wide area networks.

Improving charging capacity for wireless sensor networks by deploying one mobile vehicle with multiple removable chargers. Wireless energy transfer is a promising technology to prolong the lifetime of wireless sensor networks (WSNs), by employing charging vehicles to replenish energy to lifetime-critical sensors. Existing studies on sensor charging assumed that one or multiple charging vehicles being deployed. Such an assumption may have its limitation for a real sensor network. On one hand, it usually is insufficient to employ just one vehicle to charge many sensors in a large-scale sensor network due to the limited charging capacity of the vehicle or energy expirations of some sensors prior to the arrival of the charging vehicle. On the other hand, although the employment of multiple vehicles can significantly improve the charging capability, it is too costly in terms of the initial investment and maintenance costs on these vehicles. In this paper, we propose a novel charging model that a charging vehicle can carry multiple low-cost removable chargers and each charger is powered by a portable high-volume battery. When there are energy-critical sensors to be charged, the vehicle can carry the chargers to charge multiple sensors simultaneously, by placing one portable charger in the vicinity of one sensor. Under this novel charging model, we study the scheduling problem of the charging vehicle so that both the dead duration of sensors and the total travel distance of the mobile vehicle per tour are minimized. Since this problem is NP-hard, we instead propose a (3+ϵ)-approximation algorithm if the residual lifetime of each sensor can be ignored; otherwise, we devise a novel heuristic algorithm, where ϵ is a given constant with 0 < ϵ ≤ 1. Finally, we evaluate the performance of the proposed algorithms through experimental simulations. Experimental results show that the performance of the proposed algorithms are very promising.

Speed control of mobile chargers serving wireless rechargeable networks. Wireless rechargeable networks have attracted increasing research attention in recent years. For charging service, a mobile charger is often employed to move across the network and charge all network nodes. To reduce the charging completion time, most existing works have used the “move-then-charge” model where the charger first moves to specific spots and then starts charging nodes nearby. As a result, these works often aim to reduce the moving delay or charging delay at the spots. However, the charging opportunity on the move is largely overlooked because the charger can charge network nodes while moving, which as we analyze in this paper, has the potential to greatly reduce the charging completion time. The major challenge to exploit the charging opportunity is the setting of the moving speed of the charger. When the charger moves slow, the charging delay will be reduced (more energy will be charged during the movement) but the moving delay will increase. To deal with this challenge, we formulate the problem of delay minimization as a Traveling Salesman Problem with Speed Variations (TSP-SV) which jointly considers both charging and moving delay. We further solve the problem using linear programming to generate (1) the moving path of the charger, (2) the moving speed variations on the path and (3) the stay time at each charging spot. We also discuss possible ways to reduce the calculation complexity. Extensive simulation experiments are conducted to study the delay performance under various scenarios. The results demonstrate that our proposed method achieves much less completion time compared to the state-of-the-art work.

A Prediction-Based Charging Policy and Interference Mitigation Approach in the Wireless Powered Internet of Things The Internet of Things (IoT) technology has recently drawn more attention due to its ability to achieve the interconnections of massive physic devices. However, how to provide a reliable power supply to energy-constrained devices and improve the energy efficiency in the wireless powered IoT (WP-IoT) is a twofold challenge. In this paper, we develop a novel wireless power transmission (WPT) system, where an unmanned aerial vehicle (UAV) equipped with radio frequency energy transmitter charges the IoT devices. A machine learning framework of echo state networks together with an improved <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${k}$ </tex-math></inline-formula> -means clustering algorithm is used to predict the energy consumption and cluster all the sensor nodes at the next period, thus automatically determining the charging strategy. The energy obtained from the UAV by WPT supports the IoT devices to communicate with each other. In order to improve the energy efficiency of the WP-IoT system, the interference mitigation problem is modeled as a mean field game, where an optimal power control policy is presented to adapt and analyze the large number of sensor nodes randomly deployed in WP-IoT. The numerical results verify that our proposed dynamic charging policy effectively reduces the data packet loss rate, and that the optimal power control policy greatly mitigates the interference, and improve the energy efficiency of the whole network.

Design of Self-sustainable Wireless Sensor Networks with Energy Harvesting and Wireless Charging AbstractEnergy provisioning plays a key role in the sustainable operations of Wireless Sensor Networks (WSNs). Recent efforts deploy multi-source energy harvesting sensors to utilize ambient energy. Meanwhile, wireless charging is a reliable energy source not affected by spatial-temporal ambient dynamics. This article integrates multiple energy provisioning strategies and adaptive adjustment to accomplish self-sustainability under complex weather conditions. We design and optimize a three-tier framework with the first two tiers focusing on the planning problems of sensors with various types and distributed energy storage powered by environmental energy. Then we schedule the Mobile Chargers (MC) between different charging activities and propose an efficient, 4-factor approximation algorithm. Finally, we adaptively adjust the algorithms to capture real-time energy profiles and jointly optimize those correlated modules. Our extensive simulations demonstrate significant improvement of network lifetime (\(\)), increase of harvested energy (15%), reduction of network cost (30%), and the charging capability of MC by 100%.

GTCCS: A Game Theoretical Collaborative Charging Scheduling for On-Demand Charging Architecture. Battery energy is always limited in most wireless networked nodes, but wireless power transfer has the potential to address the problem for good. In this paper, we study the problem of collaborative charging decisions in Wireless Rechargeable Sensor Networks (WRSNs). In these networks, multiple Wireless Charging Vehicles (WCVs) need to decide which of the requesting sensors to charge and in what o...

Collaborative Mobile Charging The limited battery capacity of sensor nodes has become one of the most critical impediments that stunt the deployment of wireless sensor networks (WSNs). Recent breakthroughs in wireless energy transfer and rechargeable lithium batteries provide a promising alternative to power WSNs: mobile vehicles/robots carrying high volume batteries serve as mobile chargers to periodically deliver energy to sensor nodes. In this paper, we consider how to schedule multiple mobile chargers to optimize energy usage effectiveness, such that every sensor will not run out of energy. We introduce a novel charging paradigm, collaborative mobile charging, where mobile chargers are allowed to intentionally transfer energy between themselves. To provide some intuitive insights into the problem structure, we first consider a scenario that satisfies three conditions, and propose a scheduling algorithm, PushWait, which is proven to be optimal and can cover a one-dimensional WSN of infinite length. Then, we remove the conditions one by one, investigating chargers' scheduling in a series of scenarios ranging from the most restricted one to a general 2D WSN. Through theoretical analysis and simulations, we demonstrate the advantages of the proposed algorithms in energy usage effectiveness and charging coverage.

Space-time modeling of traffic flow. This paper discusses the application of space-time autoregressive integrated moving average (STARIMA) methodology for representing traffic flow patterns. Traffic flow data are in the form of spatial time series and are collected at specific locations at constant intervals of time. Important spatial characteristics of the space-time process are incorporated in the STARIMA model through the use of weighting matrices estimated on the basis of the distances among the various locations where data are collected. These matrices distinguish the space-time approach from the vector autoregressive moving average (VARMA) methodology and enable the model builders to control the number of the parameters that have to be estimated. The proposed models can be used for short-term forecasting of space-time stationary traffic-flow processes and for assessing the impact of traffic-flow changes on other parts of the network. The three-stage iterative space-time model building procedure is illustrated using 7.5min average traffic flow data for a set of 25 loop-detectors located at roads that direct to the centre of the city of Athens, Greece. Data for two months with different traffic-flow characteristics are modelled in order to determine the stability of the parameter estimation.

Breath: An Adaptive Protocol for Industrial Control Applications Using Wireless Sensor Networks An energy-efficient, reliable and timely data transmission is essential for Wireless Sensor Networks (WSNs) employed in scenarios where plant information must be available for control applications. To reach a maximum efficiency, cross-layer interaction is a major design paradigm to exploit the complex interaction among the layers of the protocol stack. This is challenging because latency, reliability, and energy are at odds, and resource-constrained nodes support only simple algorithms. In this paper, the novel protocol Breath is proposed for control applications. Breath is designed for WSNs where nodes attached to plants must transmit information via multihop routing to a sink. Breath ensures a desired packet delivery and delay probabilities while minimizing the energy consumption of the network. The protocol is based on randomized routing, medium access control, and duty-cycling jointly optimized for energy efficiency. The design approach relies on a constrained optimization problem, whereby the objective function is the energy consumption and the constraints are the packet reliability and delay. The challenging part is the modeling of the interactions among the layers by simple expressions of adequate accuracy, which are then used for the optimization by in-network processing. The optimal working point of the protocol is achieved by a simple algorithm, which adapts to traffic variations and channel conditions with negligible overhead. The protocol has been implemented and experimentally evaluated on a testbed with off-the-shelf wireless sensor nodes, and it has been compared with a standard IEEE 802.15.4 solution. Analytical and experimental results show that Breath is tunable and meets reliability and delay requirements. Breath exhibits a good distribution of the working load, thus ensuring a long lifetime of the network. Therefore, Breath is a good candidate for efficient, reliable, and timely data gathering for control applications.

A Survey on Vehicle Routing Problem with Loading Constraints In the classical vehicle routing problem (VRP), a fleet of vehicles is available to serve a set of customers with known demand. Each customer is visited by exactly one vehicle as well as the objective is to minimize the total distance or the total charge incurred. Recent years have seen increased attention on VRP integrated with additional loading constraints, known as 2L-CVRP or 3L-CVRP. In this paper we present a survey of the state-of-the-art on 2L-CVRP/3L-CVRP.

DeepTest: automated testing of deep-neural-network-driven autonomous cars Recent advances in Deep Neural Networks (DNNs) have led to the development of DNN-driven autonomous cars that, using sensors like camera, LiDAR, etc., can drive without any human intervention. Most major manufacturers including Tesla, GM, Ford, BMW, and Waymo/Google are working on building and testing different types of autonomous vehicles. The lawmakers of several US states including California, Texas, and New York have passed new legislation to fast-track the process of testing and deployment of autonomous vehicles on their roads. However, despite their spectacular progress, DNNs, just like traditional software, often demonstrate incorrect or unexpected corner-case behaviors that can lead to potentially fatal collisions. Several such real-world accidents involving autonomous cars have already happened including one which resulted in a fatality. Most existing testing techniques for DNN-driven vehicles are heavily dependent on the manual collection of test data under different driving conditions which become prohibitively expensive as the number of test conditions increases. In this paper, we design, implement, and evaluate DeepTest, a systematic testing tool for automatically detecting erroneous behaviors of DNN-driven vehicles that can potentially lead to fatal crashes. First, our tool is designed to automatically generated test cases leveraging real-world changes in driving conditions like rain, fog, lighting conditions, etc. DeepTest systematically explore different parts of the DNN logic by generating test inputs that maximize the numbers of activated neurons. DeepTest found thousands of erroneous behaviors under different realistic driving conditions (e.g., blurring, rain, fog, etc.) many of which lead to potentially fatal crashes in three top performing DNNs in the Udacity self-driving car challenge.

Methods for Flexible Management of Blockchain-Based Cryptocurrencies in Electricity Markets and Smart Grids The growing trend in the use of blockchain-based cryptocurrencies in modern communities provides several advantages, but also imposes several challenges to energy markets and power systems, in general. This paper aims at providing recommendations for efficient use of digital cryptocurrencies in today's and future smart power systems, in order to face the challenging aspects of this new technology. In this paper, existing issues and challenges of smart grids in the presence of blockchain-based cryptocurrencies are presented and some innovative approaches for efficient integration and management of blockchain-based cryptocurrencies in smart grids are proposed. Also some recommendations are given for improving the smart grids performance in the presence of digital cryptocurrencies and some future research directions are highlighted.

Ensuring attribute privacy protection and fast decryption for outsourced data security in mobile cloud computing. Although many users outsource their various data to clouds, data security and privacy concerns are still the biggest obstacles that hamper the widespread adoption of cloud computing. Anonymous attribute-based encryption (anonymous ABE) enables fine-grained access control over cloud storage and preserves receivers’ attribute privacy by hiding attribute information in ciphertexts. However, in existing anonymous ABE work, a user knows whether attributes and a hidden policy match or not only after repeating decryption attempts. And, each decryption usually requires many pairings and the computation overhead grows with the complexity of the access formula. Hence, existing schemes suffer a severe efficiency drawback and are not suitable for mobile cloud computing where users may be resource-constrained.

Differentially private Naive Bayes learning over multiple data sources. For meeting diverse requirements of data analysis, the machine learning classifier has been provided as a tool to evaluate data in many applications. Due to privacy concerns of preventing disclosing sensitive information, data owners often suppress their data for an untrusted trainer to train a classifier. Some existing work proposed privacy-preserving solutions for learning algorithms, which allow a trainer to build a classifier over the data from a single owner. However, they cannot be directly used in the multi-owner setting where each owner is not totally trusted for each other. In this paper, we propose a novel privacy-preserving Naive Bayes learning scheme with multiple data sources. The proposed scheme enables a trainer to train a Naive Bayes classifier over the dataset provided jointly by different data owners, without the help of a trusted curator. The training result can achieve ϵ-differential privacy while the training will not break the privacy of each owner. We implement the prototype of the scheme and conduct corresponding experiment.

Privacy-preserving communication and power injection over vehicle networks and 5G smart grid slice. As an important combination of autonomous vehicle networks (AVNs) and smart grid, the vehicle-to-grid (V2G) network can facilitate the adoption of renewable resources. Based on V2G networks, parked electric vehicles (EVs) can charge during off-peak hours and inject excess power to the grid during peak hours for earnings. However, each EV's power injection bids in V2G are sensitive and vehicle-to-vehicle (V2V) communication may be eavesdropped, which has become an obstacle to the wide deployments of AVNs. Aiming to efficiently tackle these security and privacy issues in AVNs, we propose an efficient privacy-preserving communication and power injection (ePPCP) scheme without pairings, which is suitable for vehicle networks and 5G smart grid slice. In ePPCP, each EV calculates two secret keys shared respectively by the utility company and the gateway to blind power injection bids. A novel aggregation technique called hash-then-homomorphic is used to further aggregate the blinded bids of different time slots. Our security analysis indicates that individual bids are hidden and secure V2V communication is ensured. Furthermore, extensive performance comparisons show that ePPCP is efficient in terms of the computation cost and communication overhead.

Proofs of Storage from Homomorphic Identification Protocols Proofs of storage (PoS) are interactive protocols allowing a client to verify that a server faithfully stores a file. Previous work has shown that proofs of storage can be constructed from any homomorphic linear authenticator (HLA). The latter, roughly speaking, are signature/message authentication schemes where `tags' on multiple messages can be homomorphically combined to yield a `tag' on any linear combination of these messages. We provide a framework for building public-key HLAs from any identification protocol satisfying certain homomorphic properties. We then show how to turn any public-key HLA into a publicly-verifiable PoS with communication complexity independent of the file length and supporting an unbounded number of verifications. We illustrate the use of our transformations by applying them to a variant of an identification protocol by Shoup, thus obtaining the first unbounded-use PoS based on factoring (in the random oracle model).

Revive: Rebalancing Off-Blockchain Payment Networks. Scaling the transaction throughput of decentralized blockchain ledgers such as Bitcoin and Ethereum has been an ongoing challenge. Two-party duplex payment channels have been designed and used as building blocks to construct linked payment networks, which allow atomic and trust-free payments between parties without exhausting the resources of the blockchain. Once a payment channel, however, is depleted (e.g., because transactions were mostly unidirectional) the channel would need to be closed and re-funded to allow for new transactions. Users are envisioned to entertain multiple payment channels with different entities, and as such, instead of refunding a channel (which incurs costly on-chain transactions), a user should be able to leverage his existing channels to rebalance a poorly funded channel. To the best of our knowledge, we present the first solution that allows an arbitrary set of users in a payment channel network to securely rebalance their channels, according to the preferences of the channel owners. Except in the case of disputes (similar to conventional payment channels), our solution does not require on-chain transactions and therefore increases the scalability of existing blockchains. In our security analysis, we show that an honest participant cannot lose any of its funds while rebalancing. We finally provide a proof of concept implementation and evaluation for the Ethereum network.

Anonymous attribute-based proxy re-encryption for access control in cloud computing. As a public key cryptographic primitive, attribute-based encryption ABE is promising in implementing fine-grained access control in cloud computing. However, before ABE comes into practical applications, two challenging issues have to be addressed, that is, users' attribute privacy protection and access policy update. In this paper, we tackle the aforementioned challenge for the first time by formalizing the notion of anonymous ciphertext-policy attribute-based proxy re-encryption anonymous CP-ABPRE and giving out a concrete construction. We propose a novel technique called match-then-re-encrypt, in which a matching phase is additionally introduced before the re-encryption phase. This technique uses special components of the proxy re-encryption key and ciphertext to anonymously check whether the proxy can fulfill a proxy re-encryption or not. Theoretical analysis and simulation results demonstrate that our anonymous CP-ABPRE scheme is secure and efficient. Copyright © 2016 John Wiley & Sons, Ltd.

Attribute-based encryption for fine-grained access control of encrypted data As more sensitive data is shared and stored by third-party sites on the Internet, there will be a need to encrypt data stored at these sites. One drawback of encrypting data, is that it can be selectively shared only at a coarse-grained level (i.e., giving another party your private key). We develop a new cryptosystem for fine-grained sharing of encrypted data that we call Key-Policy Attribute-Based Encryption (KP-ABE). In our cryptosystem, ciphertexts are labeled with sets of attributes and private keys are associated with access structures that control which ciphertexts a user is able to decrypt. We demonstrate the applicability of our construction to sharing of audit-log information and broadcast encryption. Our construction supports delegation of private keys which subsumesHierarchical Identity-Based Encryption (HIBE).

Image quality assessment: from error visibility to structural similarity. Objective methods for assessing perceptual image quality traditionally attempted to quantify the visibility of errors (differences) between a distorted image and a reference image using a variety of known properties of the human visual system. Under the assumption that human visual perception is highly adapted for extracting structural information from a scene, we introduce an alternative complementary framework for quality assessment based on the degradation of structural information. As a specific example of this concept, we develop a Structural Similarity Index and demonstrate its promise through a set of intuitive examples, as well as comparison to both subjective ratings and state-of-the-art objective methods on a database of images compressed with JPEG and JPEG2000.

Cell-Free Massive MIMO versus Small Cells. A Cell-Free Massive MIMO (multiple-input multiple-output) system comprises a very large number of distributed access points (APs), which simultaneously serve a much smaller number of users over the same time/frequency resources based on directly measured channel characteristics. The APs and users have only one antenna each. The APs acquire channel state information through time-division duplex operation and the reception of uplink pilot signals transmitted by the users. The APs perform multiplexing/de-multiplexing through conjugate beamforming on the downlink and matched filtering on the uplink. Closed-form expressions for individual user uplink and downlink throughputs lead to max–min power control algorithms. Max–min power control ensures uniformly good service throughout the area of coverage. A pilot assignment algorithm helps to mitigate the effects of pilot contamination, but power control is far more important in that regard. Cell-Free Massive MIMO has considerably improved performance with respect to a conventional small-cell scheme, whereby each user is served by a dedicated AP, in terms of both 95%-likely per-user throughput and immunity to shadow fading spatial correlation. Under uncorrelated shadow fading conditions, the cell-free scheme provides nearly fivefold improvement in 95%-likely per-user throughput over the small-cell scheme, and tenfold improvement when shadow fading is correlated.

Automated Synthesis of Data Paths in Digital Systems This paper presents a unifying procedure, called Facet, for the automated synthesis of data paths at the register-transfer level. The procedure minimizes the number of storage elements, data operators, and interconnection units. A design generator named Emerald, based on Facet, was developed and implemented to facilitate extensive experiments with the methodology. The input to the design generator is a behavioral description which is viewed as a code sequence. Emerald provides mechanisms for interactively manipulating the code sequence. Different forms of the code sequence are mapped into data paths of different cost and speed. Data paths for the behavioral descriptions of the AM2910, the AM2901, and the IBM System/370 were produced and analyzed. Designs for the AM2910 and the AM2901 are compared with commercial designs. Overall, the total number of gates required for Emerald's designs is about 15 percent more than the commercial designs. The design space spanned by the behavioral specification of the AM2901 is extensively explored.

Toward Social Learning Environments We are teaching a new generation of students, cradled in technologies, communication and abundance of information. The implications are that we need to focus the design of learning technologies to support social learning in context. Instead of designing technologies that “teach” the learner, the new social learning technologies will perform three main roles: 1) support the learner in finding the right content (right for the context, for the particular learner, for the specific purpose of the learner, right pedagogically); 2) support learners to connect with the right people (again right for the context, learner, purpose, educational goal etc.), and 3) motivate / incentivize people to learn. In the pursuit of such environments, new areas of sciences become relevant as a source of methods and techniques: social psychology, economic / game theory, multi-agent systems. The paper illustrates how social learning technologies can be designed using some existing and emerging technologies: ontologies vs. social tagging, exploratory search, collaborative vs. self-managed social recommendations, trust and reputation mechanisms, mechanism design and social visualization.

Robust and Imperceptible Dual Watermarking for Telemedicine Applications In this paper, the effects of different error correction codes on the robustness and imperceptibility of discrete wavelet transform and singular value decomposition based dual watermarking scheme is investigated. Text and image watermarks are embedded into cover radiological image for their potential application in secure and compact medical data transmission. Four different error correcting codes such as Hamming, the Bose, Ray-Chaudhuri, Hocquenghem (BCH), the Reed---Solomon and hybrid error correcting (BCH and repetition code) codes are considered for encoding of text watermark in order to achieve additional robustness for sensitive text data such as patient identification code. Performance of the proposed algorithm is evaluated against number of signal processing attacks by varying the strength of watermarking and covers image modalities. The experimental results demonstrate that this algorithm provides better robustness without affecting the quality of watermarked image.This algorithm combines the advantages and removes the disadvantages of the two transform techniques. Out of the three error correcting codes tested, it has been found that Reed---Solomon shows the best performance. Further, a hybrid model of two of the error correcting codes (BCH and repetition code) is concatenated and implemented. It is found that the hybrid code achieves better results in terms of robustness. This paper provides a detailed analysis of the obtained experimental results.

Automated Detection of Obstructive Sleep Apnea Events from a Single-Lead Electrocardiogram Using a Convolutional Neural Network. In this study, we propose a method for the automated detection of obstructive sleep apnea (OSA) from a single-lead electrocardiogram (ECG) using a convolutional neural network (CNN). A CNN model was designed with six optimized convolution layers including activation, pooling, and dropout layers. One-dimensional (1D) convolution, rectified linear units (ReLU), and max pooling were applied to the convolution, activation, and pooling layers, respectively. For training and evaluation of the CNN model, a single-lead ECG dataset was collected from 82 subjects with OSA and was divided into training (including data from 63 patients with 34,281 events) and testing (including data from 19 patients with 8571 events) datasets. Using this CNN model, a precision of 0.99%, a recall of 0.99%, and an F-score of 0.99% were attained with the training dataset; these values were all 0.96% when the CNN was applied to the testing dataset. These results show that the proposed CNN model can be used to detect OSA accurately on the basis of a single-lead ECG. Ultimately, this CNN model may be used as a screening tool for those suspected to suffer from OSA.

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

Deep Reinforcement Learning for Autonomous Driving: A Survey With the development of deep representation learning, the domain of reinforcement learning (RL) has become a powerful learning framework now capable of learning complex policies in high dimensional environments. This review summarises deep reinforcement learning (DRL) algorithms and provides a taxonomy of automated driving tasks where (D)RL methods have been employed, while addressing key computational challenges in real world deployment of autonomous driving agents. It also delineates adjacent domains such as behavior cloning, imitation learning, inverse reinforcement learning that are related but are not classical RL algorithms. The role of simulators in training agents, methods to validate, test and robustify existing solutions in RL are discussed.

Auto-Alert: A Spatial and Temporal Architecture for Driving Assistance in Road Traffic Environments Over the last decade, the Advanced Driver Assistance System (ADAS) concept has evolved prominently. ADAS involves several advanced approaches such as automotive electronics, vehicular communication, RADAR, LIDAR, computer vision, and its associated aspects such as machine learning and deep learning. Of these, computer vision and machine learning-based solutions have mainly been effective that have allowed real-time vehicle control, driver-aided systems, etc. However, most of the existing works deal with ADAS deployment and autonomous driving functionality in countries with well-disciplined lane traffic. These solutions and frameworks do not work in countries and cities with less-disciplined/ chaotic traffic. Hence, critical ADAS functionalities and even L2/ L3 autonomy levels in driving remain a major open challenge. In this regard, this work proposes a novel framework called Auto-Alert. Auto-Alert performs a two-stage spatial and temporal analysis based on external traffic environment and tri-axial sensor system for safe driving assistance. This work investigates time-series analysis with deep learning models for driving events prediction and assistance. Further, as a basic premise, various essential design considerations towards the ADAS are discussed. Significantly, the Convolutional Neural Network (CNN) and Long-Short-Term-Memory (LSTM) models are applied in the proposed Auto-Alert. It is shown that the LSTM outperforms the CNN with 99% for the considered window length. Importantly, this also involves developing and demonstrating an efficient traffic monitoring and density estimation system. Further, this work provides the benchmark results for Indian Driving Dataset (IDD), specifically for the object detection task. The findings of this proposed work demonstrate the significance of using CNN and LSTM networks to assist the driver in the holistic traffic environment.

Attention Based Vehicle Trajectory Prediction Self-driving vehicles need to continuously analyse the driving scene, understand the behavior of other road users and predict their future trajectories in order to plan a safe motion and reduce their reaction time. Motivated by this idea, this paper addresses the problem of vehicle trajectory prediction over an extended horizon. On highways, human drivers continuously adapt their speed and paths according to the behavior of their neighboring vehicles. Therefore, vehicles' trajectories are very correlated and considering vehicle interactions makes motion prediction possible even before the start of a clear maneuver pattern. To this end, we introduce and analyze trajectory prediction methods based on how they model the vehicles interactions. Inspired by human reasoning, we use an attention mechanism that explicitly highlights the importance of neighboring vehicles with respect to their future states. We go beyond pairwise vehicle interactions and model higher order interactions. Moreover, the existence of different goals and driving behaviors induces multiple potential futures. We exploit a combination of global and partial attention paid to surrounding vehicles to generate different possible trajectory. Experiments on highway datasets show that the proposed model outperforms the state-of-the-art performances.

Using Ontology-Based Traffic Models for More Efficient Decision Making of Autonomous Vehicles The paper describes how a high-level abstract world model can be used to support the decision-making process of an autonomous driving system. The approach uses a hierarchical world model and distinguishes between a low-level model for the trajectory planning and a high-level model for solving the traffic coordination problem. The abstract world model used in the CyberCars-2 project is presented. It is based on a topological lane segmentation and introduces relations to represent the semantic context of the traffic scenario. This makes it much easier to realize a consistent and complete driving control system, and to analyze, evaluate and simulate such a system.

Extracting Traffic Primitives Directly from Naturalistically Logged Data for Self-Driving Applications. Developing an automated vehicle, that can handle complicated driving scenarios and appropriately interact with other road users, requires the ability to semantically learn and understand driving environment, oftentimes, based on analyzing massive amounts of naturalistic driving data. An important paradigm that allows automated vehicles to both learn from human drivers and gain insights is understa...

Using Ontologies for Test Suites Generation for Automated and Autonomous Driving Functions In this paper, we outline a general automated testing approach to be applied for verification and validation of automated and autonomous driving functions. The approach makes use of ontologies of environment the system under test is interacting with. Ontologies are automatically converted into input models for combinatorial testing, which are used to generate test cases. The obtained abstract test cases are used to generate concrete test scenarios that provide the basis for simulation used to verify the functionality of the system under test. We discuss the general approach including its potential for automation in the automotive domain where there is growing need for sophisticated verification based on simulation in case of automated and autonomous vehicles.

Adversarial Evaluation of Autonomous Vehicles in Lane-Change Scenarios Autonomous vehicles must be comprehensively evaluated before deployed in cities and highways. However, most existing evaluation approaches for autonomous vehicles are static and lack adaptability, so they are usually inefficient in generating challenging scenarios for tested vehicles. In this paper, we propose an adaptive evaluation framework to efficiently evaluate autonomous vehicles in adversarial environments generated by deep reinforcement learning. Considering the multimodal nature of dangerous scenarios, we use ensemble models to represent different local optimums for diversity. We then utilize a nonparametric Bayesian method to cluster the adversarial policies. The proposed method is validated in a typical lane-change scenario that involves frequent interactions between the ego vehicle and the surrounding vehicles. Results show that the adversarial scenarios generated by our method significantly degrade the performance of the tested vehicles. We also illustrate different patterns of generated adversarial environments, which can be used to infer the weaknesses of the tested vehicles.

Decision-making in driver-automation shared control: A review and perspectives Shared control schemes allow a human driver to work with an automated driving agent in driver-vehicle systems while retaining the driver's abilities to control. The human driver, as an essential agent in the driver-vehicle shared control systems, should be precisely modeled regarding their cognitive processes, control strategies, and decision-making processes. The interactive strategy design between drivers and automated driving agents brings an excellent challenge for human-centric driver assistance systems due to the inherent characteristics of humans. Many open-ended questions arise, such as what proper role of human drivers should act in a shared control scheme? How to make an intelligent decision capable of balancing the benefits of agents in shared control systems? Due to the advent of these attentions and questions, it is desirable to present a survey on the decision making between human drivers and highly automated vehicles, to understand their architectures, human driver modeling, and interaction strategies under the driver-vehicle shared schemes. Finally, we give a further discussion on the key future challenges and opportunities. They are likely to shape new potential research directions.

A Classification Model for Sensing Human Trust in Machines Using EEG and GSR. Today, intelligent machines interact and collaborate with humans in a way that demands a greater level of trust between human and machine. A first step toward building intelligent machines that are capable of building and maintaining trust with humans is the design of a sensor that will enable machines to estimate human trust level in real time. In this article, two approaches for developing classifier-based empirical trust-sensor models are presented that specifically use electroencephalography and galvanic skin response measurements. Human subject data collected from 45 participants is used for feature extraction, feature selection, classifier training, and model validation. The first approach considers a general set of psychophysiological features across all participants as the input variables and trains a classifier-based model for each participant, resulting in a trust-sensor model based on the general feature set (i.e., a “general trust-sensor model”). The second approach considers a customized feature set for each individual and trains a classifier-based model using that feature set, resulting in improved mean accuracy but at the expense of an increase in training time. This work represents the first use of real-time psychophysiological measurements for the development of a human trust sensor. Implications of the work, in the context of trust management algorithm design for intelligent machines, are also discussed.

Hiding Traces of Resampling in Digital Images Resampling detection has become a standard tool for forensic analyses of digital images. This paper presents new variants of image transformation operations which are undetectable by resampling detectors based on periodic variations in the residual signal of local linear predictors in the spatial domain. The effectiveness of the proposed method is supported with evidence from experiments on a large image database for various parameter settings. We benchmark detectability as well as the resulting image quality against conventional linear and bicubic interpolation and interpolation with a sinc kernel. These early findings on ldquocounter-forensicrdquo techniques put into question the reliability of known forensic tools against smart counterfeiters in general, and might serve as benchmarks and motivation for the development of much improved forensic techniques.

Zee: zero-effort crowdsourcing for indoor localization Radio Frequency (RF) fingerprinting, based onWiFi or cellular signals, has been a popular approach to indoor localization. However, its adoption in the real world has been stymied by the need for sitespecific calibration, i.e., the creation of a training data set comprising WiFi measurements at known locations in the space of interest. While efforts have been made to reduce this calibration effort using modeling, the need for measurements from known locations still remains a bottleneck. In this paper, we present Zee -- a system that makes the calibration zero-effort, by enabling training data to be crowdsourced without any explicit effort on the part of users. Zee leverages the inertial sensors (e.g., accelerometer, compass, gyroscope) present in the mobile devices such as smartphones carried by users, to track them as they traverse an indoor environment, while simultaneously performing WiFi scans. Zee is designed to run in the background on a device without requiring any explicit user participation. The only site-specific input that Zee depends on is a map showing the pathways (e.g., hallways) and barriers (e.g., walls). A significant challenge that Zee surmounts is to track users without any a priori, user-specific knowledge such as the user's initial location, stride-length, or phone placement. Zee employs a suite of novel techniques to infer location over time: (a) placement-independent step counting and orientation estimation, (b) augmented particle filtering to simultaneously estimate location and user-specific walk characteristics such as the stride length,(c) back propagation to go back and improve the accuracy of ocalization in the past, and (d) WiFi-based particle initialization to enable faster convergence. We present an evaluation of Zee in a large office building.

New Stability Conditions Based on Piecewise Fuzzy Lyapunov Functions and Tensor Product Transformations Improvements of recent stability conditions for continuous-time Takagi–Sugeno (T–S) fuzzy systems are proposed. The key idea is to bring together the so-called local transformations of membership functions and new piecewise fuzzy Lyapunov functions. By relying on these special local transformations, the associated linear matrix inequalities that are used to prove the system’s stability can be relaxed without increasing the number of conditions. In addition, to enhance the usefulness of the proposed methodology, one can choose between two different sets of conditions characterized by independence or dependence on known bounds of the membership functions time derivatives. A standard example is presented to illustrate that the proposed method is able to provide substantial improvements in some cases.

Incorporation of Efficient Second-Order Solvers Into Latent Factor Models for Accurate Prediction of Missing QoS Data. Generating highly accurate predictions for missing quality-of-service (QoS) data is an important issue. Latent factor (LF)-based QoS-predictors have proven to be effective in dealing with it. However, they are based on first-order solvers that cannot well address their target problem that is inherently bilinear and nonconvex, thereby leaving a significant opportunity for accuracy improvement. This...

Blockchain-Enabled Distributed Security Framework for Next-Generation IoT: An Edge Cloud and Software-Defined Network-Integrated Approach The Internet of Things (IoT) plays a vital role in the real world by providing autonomous support for communications and operations, thus enabling and promoting novel services that are commonly used in day-to-day life. It is important to do research on security frameworks for next-generation IoT and develop state-of-the-art confidentiality protection schemes to deal with various attacks on IoT networks. In order to offer prominent features like continuous confidentiality, authentication, and robustness, the blockchain technology comes out as a sustainable solution. A blockchain-enabled distributed security framework using edge cloud and software-defined networking (SDN) is presented in this article. The security attack detection is achieved at the cloud layer, and security attacks are consequently reduced at the edge layer of the IoT network. The SDN-enabled gateway offers dynamic network traffic flow management, which contributes to the security attack recognition through determining doubtful network traffic flows and diminishes security attacks through hindering doubtful flows. The results obtained show that the proposed security framework can efficiently and effectively meet the data confidentiality challenges introduced by the integration of blockchain, edge cloud, and SDN paradigm.

Small Leak Location for Intelligent Pipeline System via Action-Dependent Heuristic Dynamic Programming In the construction process of intelligent pipeline system, pipeline monitoring is an important content to improve the safety of pipeline operation. Small leak location, in particular, is the primary focus of pipeline monitoring due to the unclear pressure drop point. To solve this, in this article, a data driven-based method of small leak location is proposed. First, through the collected pipeline parameters, empirical flow variables, and historical pressure values in pipeline system, a pipeline model based on pressure along pipeline is presented by a three-layer neural network to close to the industrial scenarios. Then, on the basis of the analyzed propagation process of negative pressure wave, an action-dependent heuristic dynamic programming with pressure–distance physical constraints is proposed to obtain the small leak location result. The proposed method is suitable for the collection of only pressure data scenario, which expands the application range. Finally, different cases of small leak location results indicate that the proposed method can locate the leak point, and the field tests further show that the proposed method has satisfactory performances in pipeline leak analysis.

The Sybil Attack Large-scale peer-to-peer systems facesecurity threats from faulty or hostile remotecomputing elements. To resist these threats, manysuch systems employ redundancy. However, if asingle faulty entity can present multiple identities,it can control a substantial fraction of the system,thereby undermining this redundancy. Oneapproach to preventing these &quot;Sybil attacks&quot; is tohave a trusted agency certify identities. Thispaper shows that, without a logically centralizedauthority, Sybil...

BLEU: a method for automatic evaluation of machine translation Human evaluations of machine translation are extensive but expensive. Human evaluations can take months to finish and involve human labor that can not be reused. We propose a method of automatic machine translation evaluation that is quick, inexpensive, and language-independent, that correlates highly with human evaluation, and that has little marginal cost per run. We present this method as an automated understudy to skilled human judges which substitutes for them when there is need for quick or frequent evaluations.

Computational thinking Summary form only given. My vision for the 21st century, Computational Thinking, will be a fundamental skill used by everyone in the world. To reading, writing, and arithmetic, we should add computational thinking to every child's analytical ability. Computational thinking involves solving problems, designing systems, and understanding human behavior by drawing on the concepts fundamental to computer science. Thinking like a computer scientist means more than being able to program a computer. It requires the ability to abstract and thus to think at multiple levels of abstraction. In this talk I will give many examples of computational thinking, argue that it has already influenced other disciplines, and promote the idea that teaching computational thinking can not only inspire future generations to enter the field of computer science but benefit people in all fields.

Fuzzy logic in control systems: fuzzy logic controller. I.

Switching between stabilizing controllers This paper deals with the problem of switching between several linear time-invariant (LTI) controllers—all of them capable of stabilizing a speci4c LTI process—in such a way that the stability of the closed-loop system is guaranteed for any switching sequence. We show that it is possible to 4nd realizations for any given family of controller transfer matrices so that the closed-loop system remains stable, no matter how we switch among the controller. The motivation for this problem is the control of complex systems where con8icting requirements make a single LTI controller unsuitable. ? 2002 Published by Elsevier Science Ltd.

Tabu Search - Part I

Bidirectional recurrent neural networks In the first part of this paper, a regular recurrent neural network (RNN) is extended to a bidirectional recurrent neural network (BRNN). The BRNN can be trained without the limitation of using input information just up to a preset future frame. This is accomplished by training it simultaneously in positive and negative time direction. Structure and training procedure of the proposed network are explained. In regression and classification experiments on artificial data, the proposed structure gives better results than other approaches. For real data, classification experiments for phonemes from the TIMIT database show the same tendency. In the second part of this paper, it is shown how the proposed bidirectional structure can be easily modified to allow efficient estimation of the conditional posterior probability of complete symbol sequences without making any explicit assumption about the shape of the distribution. For this part, experiments on real data are reported

An intensive survey of fair non-repudiation protocols With the phenomenal growth of the Internet and open networks in general, security services, such as non-repudiation, become crucial to many applications. Non-repudiation services must ensure that when Alice sends some information to Bob over a network, neither Alice nor Bob can deny having participated in a part or the whole of this communication. Therefore a fair non-repudiation protocol has to generate non-repudiation of origin evidences intended to Bob, and non-repudiation of receipt evidences destined to Alice. In this paper, we clearly define the properties a fair non-repudiation protocol must respect, and give a survey of the most important non-repudiation protocols without and with trusted third party (TTP). For the later ones we discuss the evolution of the TTP's involvement and, between others, describe the most recent protocol using a transparent TTP. We also discuss some ad-hoc problems related to the management of non-repudiation evidences.

Dynamic movement and positioning of embodied agents in multiparty conversations For embodied agents to engage in realistic multiparty conversation, they must stand in appropriate places with respect to other agents and the environment. When these factors change, such as an agent joining the conversation, the agents must dynamically move to a new location and/or orientation to accommodate. This paper presents an algorithm for simulating movement of agents based on observed human behavior using techniques developed for pedestrian movement in crowd simulations. We extend a previous group conversation simulation to include an agent motion algorithm. We examine several test cases and show how the simulation generates results that mirror real-life conversation settings.

An improved genetic algorithm with conditional genetic operators and its application to set-covering problem The genetic algorithm (GA) is a popular, biologically inspired optimization method. However, in the GA there is no rule of thumb to design the GA operators and select GA parameters. Instead, trial-and-error has to be applied. In this paper we present an improved genetic algorithm in which crossover and mutation are performed conditionally instead of probability. Because there are no crossover rate and mutation rate to be selected, the proposed improved GA can be more easily applied to a problem than the conventional genetic algorithms. The proposed improved genetic algorithm is applied to solve the set-covering problem. Experimental studies show that the improved GA produces better results over the conventional one and other methods.

Lane-level traffic estimations using microscopic traffic variables This paper proposes a novel inference method to estimate lane-level traffic flow, time occupancy and vehicle inter-arrival time on road segments where local information could not be measured and assessed directly. The main contributions of the proposed method are 1) the ability to perform lane-level estimations of traffic flow, time occupancy and vehicle inter-arrival time and 2) the ability to adapt to different traffic regimes by assessing only microscopic traffic variables. We propose a modified Kriging estimation model which explicitly takes into account both spatial and temporal variability. Performance evaluations are conducted using real-world data under different traffic regimes and it is shown that the proposed method outperforms a Kalman filter-based approach.

Convolutional Neural Network-Based Classification of Driver's Emotion during Aggressive and Smooth Driving Using Multi-Modal Camera Sensors. Because aggressive driving often causes large-scale loss of life and property, techniques for advance detection of adverse driver emotional states have become important for the prevention of aggressive driving behaviors. Previous studies have primarily focused on systems for detecting aggressive driver emotion via smart-phone accelerometers and gyro-sensors, or they focused on methods of detecting physiological signals using electroencephalography (EEG) or electrocardiogram (ECG) sensors. Because EEG and ECG sensors cause discomfort to drivers and can be detached from the driver's body, it becomes difficult to focus on bio-signals to determine their emotional state. Gyro-sensors and accelerometers depend on the performance of GPS receivers and cannot be used in areas where GPS signals are blocked. Moreover, if driving on a mountain road with many quick turns, a driver's emotional state can easily be misrecognized as that of an aggressive driver. To resolve these problems, we propose a convolutional neural network (CNN)-based method of detecting emotion to identify aggressive driving using input images of the driver's face, obtained using near-infrared (NIR) light and thermal camera sensors. In this research, we conducted an experiment using our own database, which provides a high classification accuracy for detecting driver emotion leading to either aggressive or smooth (i.e., relaxed) driving. Our proposed method demonstrates better performance than existing methods.

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

A Hybrid Architecture for Federated and Centralized Learning Many of the machine learning tasks rely on centralized learning (CL), which requires the transmission of local datasets from the clients to a parameter server (PS) entailing huge communication overhead. To overcome this, federated learning (FL) has been suggested as a promising tool, wherein the clients send only the model updates to the PS instead of the whole dataset. However, FL demands powerful computational resources from the clients. In practice, not all the clients have sufficient computational resources to participate in training. To address this common scenario, we propose a more efficient approach called hybrid federated and centralized learning (HFCL), wherein only the clients with sufficient resources employ FL, while the remaining ones send their datasets to the PS, which computes the model on behalf of them. Then, the model parameters are aggregated at the PS. To improve the efficiency of dataset transmission, we propose two different techniques: i) increased computation-per-client and ii) sequential data transmission. Notably, the HFCL frameworks outperform FL with up to 20% improvement in the learning accuracy when only half of the clients perform FL while having 50% less communication overhead than CL since all the clients collaborate on the learning process with their datasets.

The Sybil Attack Large-scale peer-to-peer systems facesecurity threats from faulty or hostile remotecomputing elements. To resist these threats, manysuch systems employ redundancy. However, if asingle faulty entity can present multiple identities,it can control a substantial fraction of the system,thereby undermining this redundancy. Oneapproach to preventing these &quot;Sybil attacks&quot; is tohave a trusted agency certify identities. Thispaper shows that, without a logically centralizedauthority, Sybil...

BLEU: a method for automatic evaluation of machine translation Human evaluations of machine translation are extensive but expensive. Human evaluations can take months to finish and involve human labor that can not be reused. We propose a method of automatic machine translation evaluation that is quick, inexpensive, and language-independent, that correlates highly with human evaluation, and that has little marginal cost per run. We present this method as an automated understudy to skilled human judges which substitutes for them when there is need for quick or frequent evaluations.

Computational thinking Summary form only given. My vision for the 21st century, Computational Thinking, will be a fundamental skill used by everyone in the world. To reading, writing, and arithmetic, we should add computational thinking to every child's analytical ability. Computational thinking involves solving problems, designing systems, and understanding human behavior by drawing on the concepts fundamental to computer science. Thinking like a computer scientist means more than being able to program a computer. It requires the ability to abstract and thus to think at multiple levels of abstraction. In this talk I will give many examples of computational thinking, argue that it has already influenced other disciplines, and promote the idea that teaching computational thinking can not only inspire future generations to enter the field of computer science but benefit people in all fields.

Fuzzy logic in control systems: fuzzy logic controller. I.

Switching between stabilizing controllers This paper deals with the problem of switching between several linear time-invariant (LTI) controllers—all of them capable of stabilizing a speci4c LTI process—in such a way that the stability of the closed-loop system is guaranteed for any switching sequence. We show that it is possible to 4nd realizations for any given family of controller transfer matrices so that the closed-loop system remains stable, no matter how we switch among the controller. The motivation for this problem is the control of complex systems where con8icting requirements make a single LTI controller unsuitable. ? 2002 Published by Elsevier Science Ltd.

Tabu Search - Part I

Bidirectional recurrent neural networks In the first part of this paper, a regular recurrent neural network (RNN) is extended to a bidirectional recurrent neural network (BRNN). The BRNN can be trained without the limitation of using input information just up to a preset future frame. This is accomplished by training it simultaneously in positive and negative time direction. Structure and training procedure of the proposed network are explained. In regression and classification experiments on artificial data, the proposed structure gives better results than other approaches. For real data, classification experiments for phonemes from the TIMIT database show the same tendency. In the second part of this paper, it is shown how the proposed bidirectional structure can be easily modified to allow efficient estimation of the conditional posterior probability of complete symbol sequences without making any explicit assumption about the shape of the distribution. For this part, experiments on real data are reported

An intensive survey of fair non-repudiation protocols With the phenomenal growth of the Internet and open networks in general, security services, such as non-repudiation, become crucial to many applications. Non-repudiation services must ensure that when Alice sends some information to Bob over a network, neither Alice nor Bob can deny having participated in a part or the whole of this communication. Therefore a fair non-repudiation protocol has to generate non-repudiation of origin evidences intended to Bob, and non-repudiation of receipt evidences destined to Alice. In this paper, we clearly define the properties a fair non-repudiation protocol must respect, and give a survey of the most important non-repudiation protocols without and with trusted third party (TTP). For the later ones we discuss the evolution of the TTP's involvement and, between others, describe the most recent protocol using a transparent TTP. We also discuss some ad-hoc problems related to the management of non-repudiation evidences.

Dynamic movement and positioning of embodied agents in multiparty conversations For embodied agents to engage in realistic multiparty conversation, they must stand in appropriate places with respect to other agents and the environment. When these factors change, such as an agent joining the conversation, the agents must dynamically move to a new location and/or orientation to accommodate. This paper presents an algorithm for simulating movement of agents based on observed human behavior using techniques developed for pedestrian movement in crowd simulations. We extend a previous group conversation simulation to include an agent motion algorithm. We examine several test cases and show how the simulation generates results that mirror real-life conversation settings.

An improved genetic algorithm with conditional genetic operators and its application to set-covering problem The genetic algorithm (GA) is a popular, biologically inspired optimization method. However, in the GA there is no rule of thumb to design the GA operators and select GA parameters. Instead, trial-and-error has to be applied. In this paper we present an improved genetic algorithm in which crossover and mutation are performed conditionally instead of probability. Because there are no crossover rate and mutation rate to be selected, the proposed improved GA can be more easily applied to a problem than the conventional genetic algorithms. The proposed improved genetic algorithm is applied to solve the set-covering problem. Experimental studies show that the improved GA produces better results over the conventional one and other methods.

Lane-level traffic estimations using microscopic traffic variables This paper proposes a novel inference method to estimate lane-level traffic flow, time occupancy and vehicle inter-arrival time on road segments where local information could not be measured and assessed directly. The main contributions of the proposed method are 1) the ability to perform lane-level estimations of traffic flow, time occupancy and vehicle inter-arrival time and 2) the ability to adapt to different traffic regimes by assessing only microscopic traffic variables. We propose a modified Kriging estimation model which explicitly takes into account both spatial and temporal variability. Performance evaluations are conducted using real-world data under different traffic regimes and it is shown that the proposed method outperforms a Kalman filter-based approach.

Convolutional Neural Network-Based Classification of Driver's Emotion during Aggressive and Smooth Driving Using Multi-Modal Camera Sensors. Because aggressive driving often causes large-scale loss of life and property, techniques for advance detection of adverse driver emotional states have become important for the prevention of aggressive driving behaviors. Previous studies have primarily focused on systems for detecting aggressive driver emotion via smart-phone accelerometers and gyro-sensors, or they focused on methods of detecting physiological signals using electroencephalography (EEG) or electrocardiogram (ECG) sensors. Because EEG and ECG sensors cause discomfort to drivers and can be detached from the driver's body, it becomes difficult to focus on bio-signals to determine their emotional state. Gyro-sensors and accelerometers depend on the performance of GPS receivers and cannot be used in areas where GPS signals are blocked. Moreover, if driving on a mountain road with many quick turns, a driver's emotional state can easily be misrecognized as that of an aggressive driver. To resolve these problems, we propose a convolutional neural network (CNN)-based method of detecting emotion to identify aggressive driving using input images of the driver's face, obtained using near-infrared (NIR) light and thermal camera sensors. In this research, we conducted an experiment using our own database, which provides a high classification accuracy for detecting driver emotion leading to either aggressive or smooth (i.e., relaxed) driving. Our proposed method demonstrates better performance than existing methods.

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

RGB-D sensing based human action and interaction analysis: a survey •A thorough overview of human action and interaction recognition using RGB-D sensors is presented.•A comprehensive analysis of both hand-crafted and deep learning based methods is conducted.•The challenges of human activity recognition using RGB-D data and existing solutions are discussed.

A predictive controller for autonomous vehicle path tracking This paper presents a model predictive controller (MPC) structure for solving the path-tracking problem of terrestrial autonomous vehicles. To achieve the desired performance during high-speed driving, the controller architecture considers both the kinematic and the dynamic control in a cascade structure. Our study contains a comparative study between two kinematic linear predictive control strategies: The first strategy is based on the successive linearization concept, and the other strategy combines a local reference frame with an approaching path strategy. Our goal is to search for the strategy that best comprises the performance and hardware-cost criteria. For the dynamic controller, a decentralized predictive controller based on a linearized model of the vehicle is used. Practical experiments obtained using an autonomous "Mini-Baja" vehicle equipped with an embedded computing system are presented. These results confirm that the proposed MPC structure is the solution that better matches the target criteria.

An Efficient Visibility Enhancement Algorithm for Road Scenes Captured by Intelligent Transportation Systems The visibility of images of outdoor road scenes will generally become degraded when captured during inclement weather conditions. Drivers often turn on the headlights of their vehicles and streetlights are often activated, resulting in localized light sources in images capturing road scenes in these conditions. Additionally, sandstorms are also weather events that are commonly encountered when driving in some regions. In sandstorms, atmospheric sand has a propensity to irregularly absorb specific portions of a spectrum, thereby causing color-shift problems in the captured image. Traditional state-of-the-art restoration techniques are unable to effectively cope with these hazy road images that feature localized light sources or color-shift problems. In response, we present a novel and effective haze removal approach to remedy problems caused by localized light sources and color shifts, which thereby achieves superior restoration results for single hazy images. The performance of the proposed method has been proven through quantitative and qualitative evaluations. Experimental results demonstrate that the proposed haze removal technique can more effectively recover scene radiance while demanding fewer computational costs than traditional state-of-the-art haze removal techniques.

Automatic Detection and Classification of Road Lane Markings Using Onboard Vehicular Cameras This paper presents a new approach for road lane classification using an onboard camera. Initially, lane boundaries are detected using a linear–parabolic lane model, and an automatic on-the-fly camera calibration procedure is applied. Then, an adaptive smoothing scheme is applied to reduce noise while keeping close edges separated, and pairs of local maxima–minima of the gradient are used as cues to identify lane markings. Finally, a Bayesian classifier based on mixtures of Gaussians is applied to classify the lane markings present at each frame of a video sequence as dashed, solid, dashed solid, solid dashed, or double solid. Experimental results indicate an overall accuracy of over 96% using a variety of video sequences acquired with different devices and resolutions.

Pedestrian Tracking Using Online Boosted Random Ferns Learning in Far-Infrared Imagery for Safe Driving at Night. Pedestrian-vehicle accidents that occur at night are a major social problem worldwide. Advanced driver assistance systems that are equipped with cameras have been designed to automatically prevent such accidents. Among the various types of cameras used in such systems, far-infrared (FIR) cameras are favorable because they are invariant to illumination changes. Therefore, this paper focuses on a pedestrian nighttime tracking system with an FIR camera that is able to discern thermal energy and is mounted on the forward roof part of a vehicle. Since the temperature difference between the pedestrian and background depends on the season and the weather, we therefore propose two models to detect pedestrians according to the season and the weather, which are determined using Weber–Fechner's law. For tracking pedestrians, we perform real-time online learning to track pedestrians using boosted random ferns and update the trackers at each frame. In particular, we link detection responses to trajectories based on similarities in position, size, and appearance. There is no standard data set for evaluating the tracking performance using an FIR camera; thus, we created the Keimyung University tracking data set (KMUTD) by combining the KMU sudden pedestrian crossing (SPC) data set [21] for summer nights with additional tracking data for winter nights. The KMUTD contains video sequences involving a moving camera, moving pedestrians, sudden shape deformations, unexpected motion changes, and partial or full occlusions between pedestrians at night. The proposed algorithm is successfully applied to various pedestrian video sequences of the KMUTD; specifically, the proposed algorithm yields more accurate tracking performance than other existing methods.

Effects of Different Alcohol Dosages on Steering Behavior in Curve Driving. Objective: The aim of this article is to explore the detailed characteristics of steering behavior in curve driving at different alcohol dosages. Background: Improper operation of the steering wheel is a contributing factor to increased crash risks on curves. Method: The experiments were conducted using a driving simulator. Twenty-five licensed drivers were recruited to perform the experiments at the four different breath alcohol concentration (BrAC) levels. The steering angle (SA), steering speed (SS), steering reversal rate (SRR), and peak-to-peak value of the steering angle (PP) were used to characterize the steering behavior. The vehicle's speed and the number of lane exceedances per kilometer were also used to examine the driving performance. Results: The SSs on the 200 m (chi(2)(3) = 20.67, p < .001), 500 m (chi(2)(3) = 22.42, p < .001), and 800 m (chi(2)(3) = 22.86, p < .001) radius curves were significantly faster for drivers under the influence of alcohol compared with those given a placebo. There were significant effects of alcohol on the SRR and PP on the 200 m, 500 m, and 800 m radius curves. Conclusion: For all of the curves, the SS, SRR, and PP had a tendency to increase as the BrAC increased. The large PP at a high BrAC, accompanied by the high speed, SS, and SRR, resulted in a high probability of lane exceedance. The use of measures of SS, SRR, and PP aided in the improvement of the accuracy of the intoxication detection for the different types of curves. Application: The most important application is to provide guidance for detecting alcohol-impaired-driving.

Risky Driver Recognition Based on Vehicle Speed Time Series. Risky driving is a major cause of traffic accidents. In this paper, we propose a new method that recognizes risky driving behaviors purely based on vehicle speed time series. This method first retrieves the important distribution pattern of the sampled positive speed-change (value and duration) tuples for individual drivers within different speed ranges. Then, it identifies the risky drivers based...

Driver Fatigue Detection Systems: A Review Driver fatigue has been attributed to traffic accidents; therefore, fatigue-related traffic accidents have a higher fatality rate and cause more damage to the surroundings compared with accidents where the drivers are alert. Recently, many automobile companies have installed driver assistance technologies in vehicles for driver assistance. Third party companies are also manufacturing fatigue detection devices; however, much research is still required for improvement. In the field of driver fatigue detection, continuous research is being performed and several articles propose promising results in constrained environments, still much progress is required. This paper presents state-of-the-art review of recent advancement in the field of driver fatigue detection. Methods are categorized into five groups, i.e., subjective reporting, driver biological features, driver physical features, vehicular features while driving, and hybrid features depending on the features used for driver fatigue detection. Various approaches have been compared for fatigue detection, and areas open for improvements are deduced.

The ApolloScape Dataset for Autonomous Driving Scene parsing aims to assign a class (semantic) label for each pixel in an image. It is a comprehensive analysis of an image. Given the rise of autonomous driving, pixel-accurate environmental perception is expected to be a key enabling technical piece. However, providing a large scale dataset for the design and evaluation of scene parsing algorithms, in particular for outdoor scenes, has been difficult. The per-pixel labelling process is prohibitively expensive, limiting the scale of existing ones. In this paper, we present a large-scale open dataset, ApolloScape, that consists of RGB videos and corresponding dense 3D point clouds. Comparing with existing datasets, our dataset has the following unique properties. The first is its scale, our initial release contains over 140K images - each with its per-pixel semantic mask, up to 1M is scheduled. The second is its complexity. Captured in various traffic conditions, the number of moving objects averages from tens to over one hundred (Figure 1). And the third is the 3D attribute, each image is tagged with high-accuracy pose information at cm accuracy and the static background point cloud has mm relative accuracy. We are able to label these many images by an interactive and efficient labelling pipeline that utilizes the high-quality 3D point cloud. Moreover, our dataset also contains different lane markings based on the lane colors and styles. We expect our new dataset can deeply benefit various autonomous driving related applications that include but not limited to 2D/3D scene understanding, localization, transfer learning, and driving simulation.

Online Palmprint Identification Biometrics-based personal identification is regarded as an effective method for automatically recognizing, with a high confidence, a person's identity. This paper presents a new biometric approach to online personal identification using palmprint technology. In contrast to the existing methods, our online palmprint identification system employs low-resolution palmprint images to achieve effective personal identification. The system consists of two parts: a novel device for online palmprint image acquisition and an efficient algorithm for fast palmprint recognition. A robust image coordinate system is defined to facilitate image alignment for feature extraction. In addition, a 2D Gabor phase encoding scheme is proposed for palmprint feature extraction and representation. The experimental results demonstrate the feasibility of the proposed system.

Touch Is Everywhere: Floor Surfaces as Ambient Haptic Interfaces Floor surfaces are notable for the diverse roles that they play in our negotiation of everyday environments. Haptic communication via floor surfaces could enhance or enable many computer-supported activities that involve movement on foot. In this paper, we discuss potential applications of such interfaces in everyday environments and present a haptically augmented floor component through which several interaction methods are being evaluated. We describe two approaches to the design of structured vibrotactile signals for this device. The first is centered on a musical phrase metaphor, as employed in prior work on tactile display. The second is based upon the synthesis of rhythmic patterns of virtual physical impact transients. We report on an experiment in which participants were able to identify communication units that were constructed from these signals and displayed via a floor interface at well above chance levels. The results support the feasibility of tactile information display via such interfaces and provide further indications as to how to effectively design vibrotactile signals for them.

Multimodal Feature-Based Surface Material Classification. When a tool is tapped on or dragged over an object surface, vibrations are induced in the tool, which can be captured using acceleration sensors. The tool-surface interaction additionally creates audible sound waves, which can be recorded using microphones. Features extracted from camera images provide additional information about the surfaces. We present an approach for tool-mediated surface clas...

Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecasting. Timely accurate traffic forecast is crucial for urban traffic control and guidance. Due to the high nonlinearity and complexity of traffic flow, traditional methods cannot satisfy the requirements of mid-and-long term prediction tasks and often neglect spatial and temporal dependencies. In this paper, we propose a novel deep learning framework, Spatio-Temporal Graph Convolutional Networks (STGCN), to tackle the time series prediction problem in traffic domain. Instead of applying regular convolutional and recurrent units, we formulate the problem on graphs and build the model with complete convolutional structures, which enable much faster training speed with fewer parameters. Experiments show that our model STGCN effectively captures comprehensive spatio-temporal correlations through modeling multi-scale traffic networks and consistently outperforms state-of-the-art baselines on various real-world traffic datasets.

Sustainable and Efficient Data Collection from WSNs to Cloud. The development of cloud computing pours great vitality into traditional wireless sensor networks (WSNs). The integration of WSNs and cloud computing has received a lot of attention from both academia and industry. However, collecting data from WSNs to cloud is not sustainable. Due to the weak communication ability of WSNs, uploading big sensed data to the cloud within the limited time becomes a b...

DeepMove: Predicting Human Mobility with Attentional Recurrent Networks. Human mobility prediction is of great importance for a wide spectrum of location-based applications. However, predicting mobility is not trivial because of three challenges: 1) the complex sequential transition regularities exhibited with time-dependent and high-order nature; 2) the multi-level periodicity of human mobility; and 3) the heterogeneity and sparsity of the collected trajectory data. In this paper, we propose DeepMove, an attentional recurrent network for mobility prediction from lengthy and sparse trajectories. In DeepMove, we first design a multi-modal embedding recurrent neural network to capture the complicated sequential transitions by jointly embedding the multiple factors that govern the human mobility. Then, we propose a historical attention model with two mechanisms to capture the multi-level periodicity in a principle way, which effectively utilizes the periodicity nature to augment the recurrent neural network for mobility prediction. We perform experiments on three representative real-life mobility datasets, and extensive evaluation results demonstrate that our model outperforms the state-of-the-art models by more than 10%. Moreover, compared with the state-of-the-art neural network models, DeepMove provides intuitive explanations into the prediction and sheds light on interpretable mobility prediction.

Modeling Location-Based User Rating Profiles for Personalized Recommendation This article proposes LA-LDA, a location-aware probabilistic generative model that exploits location-based ratings to model user profiles and produce recommendations. Most of the existing recommendation models do not consider the spatial information of users or items; however, LA-LDA supports three classes of location-based ratings, namely spatial user ratings for nonspatial items, nonspatial user ratings for spatial items, and spatial user ratings for spatial items. LA-LDA consists of two components, ULA-LDA and ILA-LDA, which are designed to take into account user and item location information, respectively. The component ULA-LDA explicitly incorporates and quantifies the influence from local public preferences to produce recommendations by considering user home locations, whereas the component ILA-LDA recommends items that are closer in both taste and travel distance to the querying users by capturing item co-occurrence patterns, as well as item location co-occurrence patterns. The two components of LA-LDA can be applied either separately or collectively, depending on the available types of location-based ratings. To demonstrate the applicability and flexibility of the LA-LDA model, we deploy it to both top-k recommendation and cold start recommendation scenarios. Experimental evidence on large-scale real-world data, including the data from Gowalla (a location-based social network), DoubanEvent (an event-based social network), and MovieLens (a movie recommendation system), reveal that LA-LDA models user profiles more accurately by outperforming existing recommendation models for top-k recommendation and the cold start problem.

Multi-Task Travel Route Planning With a Flexible Deep Learning Framework Travel route planning aims to map out a feasible sightseeing itinerary for a traveler covering famous attractions and meeting the tourist’s desire. It is very useful for tourists to plan their travel routes when they want to travel at unfamiliar scenic cities. Existing methods for travel route planning mainly concentrate on a single planning problem for a special task, but is not capable of being ...

Deep Irregular Convolutional Residual LSTM for Urban Traffic Passenger Flows Prediction Urban traffic passenger flows prediction is practically important to facilitate many real applications including transportation management and public safety. Recently, deep learning based approaches are proposed to learn the spatio-temporal characteristics of the traffic passenger flows. However, it is still very challenging to handle some complex factors such as hybrid transportation lines, mixed traffic, transfer stations, and some extreme weathers. Considering the multi-channel and irregularity properties of urban traffic passenger flows in different transportation lines, a more efficient and fine-grained deep spatio-temporal feature learning model is necessary. In this paper, we propose a deep irregular convolutional residual LSTM network model called DST-ICRL for urban traffic passenger flows prediction. We first model the passenger flows among different traffic lines in a transportation network into multi-channel matrices analogous to the RGB pixel matrices of an image. Then, we propose a deep learning framework that integrates irregular convolutional residential network and LSTM units to learn the spatial-temporal feature representations. To fully utilize the historical passenger flows, we sample both the short-term and long-term historical traffic data, which can capture the periodicity and trend of the traffic passenger flows. In addition, we also fuse other external factors further to facilitate a real-time prediction. We conduct extensive experiments on different types of traffic passenger flows datasets including subway, taxi and bus flows in Beijing as well as bike flows in New York. The results show that the proposed DST-ICRL significantly outperforms both traditional and deep learning based urban traffic passenger flows prediction methods.

Spatial Temporal Incidence Dynamic Graph Neural Networks for Traffic Flow Forecasting •The paper proposes a novel dynamic graph recurrent convolutional neural network model, named Dynamic-GRCNN, to deeply capture the spatio-temporal traffic flow features for more accurately predicting urban passenger traffic flows.•The paper presents incidence dynamic graph structures based on historically passenger traffic flows to model traffic station relationships. Different from existing traffic transportation network topological structures based graph relationships between stations, the incidence dynamic graph structures firstly model the traffic relationships from historical passenger flows.•For real urban passenger traffic flows, the paper demonstrates that dynamic spatial-temporal incidence graphs are more suitable to model external changes and influences.•The paper compares Dynamic-GRCNN with state-of-the-art deep learning approaches on three benchmark datasets which contain different types of passenger traffic flows for evaluation. The results show that Dynamic-GRCNN significantly outperforms all the baselines in both effectiveness and efficiency in urban passenger traffic flows prediction.

Distinctive Image Features from Scale-Invariant Keypoints This paper presents a method for extracting distinctive invariant features from images that can be used to perform reliable matching between different views of an object or scene. The features are invariant to image scale and rotation, and are shown to provide robust matching across a substantial range of affine distortion, change in 3D viewpoint, addition of noise, and change in illumination. The features are highly distinctive, in the sense that a single feature can be correctly matched with high probability against a large database of features from many images. This paper also describes an approach to using these features for object recognition. The recognition proceeds by matching individual features to a database of features from known objects using a fast nearest-neighbor algorithm, followed by a Hough transform to identify clusters belonging to a single object, and finally performing verification through least-squares solution for consistent pose parameters. This approach to recognition can robustly identify objects among clutter and occlusion while achieving near real-time performance.

Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation Object detection performance, as measured on the canonical PASCAL VOC dataset, has plateaued in the last few years. The best-performing methods are complex ensemble systems that typically combine multiple low-level image features with high-level context. In this paper, we propose a simple and scalable detection algorithm that improves mean average precision (mAP) by more than 30% relative to the previous best result on VOC 2012 -- achieving a mAP of 53.3%. Our approach combines two key insights: (1) one can apply high-capacity convolutional neural networks (CNNs) to bottom-up region proposals in order to localize and segment objects and (2) when labeled training data is scarce, supervised pre-training for an auxiliary task, followed by domain-specific fine-tuning, yields a significant performance boost. Since we combine region proposals with CNNs, we call our method R-CNN: Regions with CNN features. We also present experiments that provide insight into what the network learns, revealing a rich hierarchy of image features. Source code for the complete system is available at http://www.cs.berkeley.edu/~rbg/rcnn.

New approach using ant colony optimization with ant set partition for fuzzy control design applied to the ball and beam system. In this paper we describe the design of a fuzzy logic controller for the ball and beam system using a modified Ant Colony Optimization (ACO) method for optimizing the type of membership functions, the parameters of the membership functions and the fuzzy rules. This is achieved by applying a systematic and hierarchical optimization approach modifying the conventional ACO algorithm using an ant set partition strategy. The simulation results show that the proposed algorithm achieves better results than the classical ACO algorithm for the design of the fuzzy controller.

Image analogies This paper describes a new framework for processing images by example, called “image analogies.” The framework involves two stages: a design phase, in which a pair of images, with one image purported to be a “filtered” version of the other, is presented as “training data”; and an application phase, in which the learned filter is applied to some new target image in order to create an “analogous” filtered result. Image analogies are based on a simple multi-scale autoregression, inspired primarily by recent results in texture synthesis. By choosing different types of source image pairs as input, the framework supports a wide variety of “image filter” effects, including traditional image filters, such as blurring or embossing; improved texture synthesis, in which some textures are synthesized with higher quality than by previous approaches; super-resolution, in which a higher-resolution image is inferred from a low-resolution source; texture transfer, in which images are “texturized” with some arbitrary source texture; artistic filters, in which various drawing and painting styles are synthesized based on scanned real-world examples; and texture-by-numbers, in which realistic scenes, composed of a variety of textures, are created using a simple painting interface.

Noninterference for a Practical DIFC-Based Operating System The Flume system is an implementation of decentralized information flow control (DIFC) at the operating system level. Prior work has shown Flume can be implemented as a practical extension to the Linux operating system, allowing real Web applications to achieve useful security guarantees. However, the question remains if the Flume system is actually secure. This paper compares Flume with other recent DIFC systems like Asbestos, arguing that the latter is inherently susceptible to certain wide-bandwidth covert channels, and proving their absence in Flume by means of a noninterference proof in the communicating sequential processes formalism.

Lower Extremity Exoskeletons and Active Orthoses: Challenges and State-of-the-Art In the nearly six decades since researchers began to explore methods of creating them, exoskeletons have progressed from the stuff of science fiction to nearly commercialized products. While there are still many challenges associated with exoskeleton development that have yet to be perfected, the advances in the field have been enormous. In this paper, we review the history and discuss the state-of-the-art of lower limb exoskeletons and active orthoses. We provide a design overview of hardware, actuation, sensory, and control systems for most of the devices that have been described in the literature, and end with a discussion of the major advances that have been made and hurdles yet to be overcome.

Biologically-inspired soft exosuit. In this paper, we present the design and evaluation of a novel soft cable-driven exosuit that can apply forces to the body to assist walking. Unlike traditional exoskeletons which contain rigid framing elements, the soft exosuit is worn like clothing, yet can generate moments at the ankle and hip with magnitudes of 18% and 30% of those naturally generated by the body during walking, respectively. Our design uses geared motors to pull on Bowden cables connected to the suit near the ankle. The suit has the advantages over a traditional exoskeleton in that the wearer's joints are unconstrained by external rigid structures, and the worn part of the suit is extremely light, which minimizes the suit's unintentional interference with the body's natural biomechanics. However, a soft suit presents challenges related to actuation force transfer and control, since the body is compliant and cannot support large pressures comfortably. We discuss the design of the suit and actuation system, including principles by which soft suits can transfer force to the body effectively and the biological inspiration for the design. For a soft exosuit, an important design parameter is the combined effective stiffness of the suit and its interface to the wearer. We characterize the exosuit's effective stiffness, and present preliminary results from it generating assistive torques to a subject during walking. We envision such an exosuit having broad applicability for assisting healthy individuals as well as those with muscle weakness.

Robust Sparse Linear Discriminant Analysis Linear discriminant analysis (LDA) is a very popular supervised feature extraction method and has been extended to different variants. However, classical LDA has the following problems: 1) The obtained discriminant projection does not have good interpretability for features. 2) LDA is sensitive to noise. 3) LDA is sensitive to the selection of number of projection directions. In this paper, a novel feature extraction method called robust sparse linear discriminant analysis (RSLDA) is proposed to solve the above problems. Specifically, RSLDA adaptively selects the most discriminative features for discriminant analysis by introducing the l2;1 norm. An orthogonal matrix and a sparse matrix are also simultaneously introduced to guarantee that the extracted features can hold the main energy of the original data and enhance the robustness to noise, and thus RSLDA has the potential to perform better than other discriminant methods. Extensive experiments on six databases demonstrate that the proposed method achieves the competitive performance compared with other state-of-the-art feature extraction methods. Moreover, the proposed method is robust to the noisy data. IEEE

Ethical Considerations Of Applying Robots In Kindergarten Settings: Towards An Approach From A Macroperspective In child-robot interaction (cHRI) research, many studies pursue the goal to develop interactive systems that can be applied in everyday settings. For early education, increasingly, the setting of a kindergarten is targeted. However, when cHRI and research are brought into a kindergarten, a range of ethical and related procedural aspects have to be considered and dealt with. While ethical models elaborated within other human-robot interaction settings, e.g., assisted living contexts, can provide some important indicators for relevant issues, we argue that it is important to start developing a systematic approach to identify and tackle those ethical issues which rise with cHRI in kindergarten settings on a more global level and address the impact of the technology from a macroperspective beyond the effects on the individual. Based on our experience in conducting studies with children in general and pedagogical considerations on the role of the institution of kindergarten in specific, in this paper, we enfold some relevant aspects that have barely been addressed in an explicit way in current cHRI research. Four areas are analyzed and key ethical issues are identified in each area: (1) the institutional setting of a kindergarten, (2) children as a vulnerable group, (3) the caregivers' role, and (4) pedagogical concepts. With our considerations, we aim at (i) broadening the methodology of the current studies within the area of cHRI, (ii) revalidate it based on our comprehensive empirical experience with research in kindergarten settings, both laboratory and real-world contexts, and (iii) provide a framework for the development of a more systematic approach to address the ethical issues in cHRI research within kindergarten settings.

ACN-Data - Analysis and Applications of an Open EV Charging Dataset. We are releasing ACN-Data, a dynamic dataset of workplace EV charging which currently includes over 30,000 sessions with more added daily. In this paper we describe the dataset, as well as some interesting user behavior it exhibits. To demonstrate the usefulness of the dataset, we present three examples, learning and predicting user behavior using Gaussian mixture models, optimally sizing on-site solar generation for adaptive electric vehicle charging, and using workplace charging to smooth the net demand Duck Curve.

Smoothed Least-laxity-first Algorithm for EV Charging. We formulate EV charging as a feasibility problem that meets all EVs' energy demands before departure under charging rate constraints and total power constraint. We propose an online algorithm, the smoothed least-laxity-first (sLLF) algorithm, that decides on the current charging rates based on only the information up to the current time. We characterize the performance of the sLLF algorithm analytically and numerically. Numerical experiments with real-world data show that it has significantly higher rate of generating feasible EV charging than several other common EV charging algorithms.

Robust Online Algorithms for Peak-Minimizing EV Charging Under Multistage Uncertainty. In this paper, we study how to utilize forecasts to design online electrical vehicle (EV) charging algorithms that can attain strong performance guarantees. We consider the scenario of an aggregator serving a large number of EVs together with its background load, using both its own renewable energy (for free) and the energy procured from the external grid. The goal of the aggregator is to minimize its peak procurement from the grid, subject to the constraint that each EV has to be fully charged before its deadline. Further, the aggregator can predict the future demand and the renewable energy supply with some levels of uncertainty. We show that such prediction can be very effective in reducing the competitive ratios of online control algorithms, and even allow online algorithms to achieve close-to-offline-optimal peak. Specifically, we first propose a 2-level increasing precision model (2-IPM), to model forecasts with different levels of accuracy. We then develop a powerful computational approach that can compute the optimal competitive ratio under 2-IPM over any online algorithm, and also online algorithms that can achieve the optimal competitive ratio. Simulation results show that, even with up to 20% day-ahead prediction errors, our online algorithms still achieve competitive ratios fairly close to 1, which are much better than the classic results in the literature with a competitive ratio of e. The second contribution of this paper is that we solve a dilemma for online algorithm design, e.g., an online algorithm with good competitive ratio may exhibit poor average-case performance. We propose a new Algorithm-Robustification procedure that can convert an online algorithm with good average-case performance to one with both the optimal competitive ratio and good average-case performance. We demonstrate via trace-based simulations the superior performance of the robustified version of a well-known heuristic algorithm based on model predictive control.

Multi-Tenant Cross-Slice Resource Orchestration: A Deep Reinforcement Learning Approach. In a software-defined radio access network (RAN), a major challenge lies in how to support diverse services for mobile users (MUs) over a common physical network infrastructure. Network slicing is a promising solution to tailor the network to match such service requests. This paper considers a software-defined RAN, where a limited number of channels are auctioned across scheduling slots to MUs of multiple service providers (SPs) (i.e., the tenants). Each SP behaves selfishly to maximize the expected long-term payoff from competition with other SPs for the orchestration of channel access opportunities over its MUs, which request both mobile-edge computing and traditional cellular services in the slices. This problem is modelled as a stochastic game, in which the decision makings of a SP depend on the network dynamics as well as the control policies of its competitors. We propose an abstract stochastic game to approximate the Nash equilibrium. The selfish behaviours of a SP can then be characterized by a single-agent Markov decision process (MDP). To simplify decision makings, we linearly decompose the per-SP MDP and derive an online scheme based on deep reinforcement learning to approach the optimal abstract control policies. Numerical experiments show significant performance gains from our scheme.

Competitive Algorithms for the Online Multiple Knapsack Problem with Application to Electric Vehicle Charging AbstractWe introduce and study a general version of the fractional online knapsack problem with multiple knapsacks, heterogeneous constraints on which items can be assigned to which knapsack, and rate-limiting constraints on the assignment of items to knapsacks. This problem generalizes variations of the knapsack problem and of the one-way trading problem that have previously been treated separately, and additionally finds application to the real-time control of electric vehicle (EV) charging. We introduce a new algorithm that achieves a competitive ratio within an additive factor of one of the best achievable competitive ratios for the general problem and matches or improves upon the best-known competitive ratio for special cases in the knapsack and one-way trading literatures. Moreover, our analysis provides a novel approach to online algorithm design based on an instance-dependent primal-dual analysis that connects the identification of worst-case instances to the design of algorithms. Finally, we illustrate the proposed algorithm via trace-based experiments of EV charging.

Hierarchical Electric Vehicle Charging Aggregator Strategy Using Dantzig-Wolfe Decomposition. This article focuses on reducing a charging cost for electric vehicles (EVs). A charging strategy is proposed to minimize the charging cost of EVs within the charging station constraints.

The Sybil Attack Large-scale peer-to-peer systems facesecurity threats from faulty or hostile remotecomputing elements. To resist these threats, manysuch systems employ redundancy. However, if asingle faulty entity can present multiple identities,it can control a substantial fraction of the system,thereby undermining this redundancy. Oneapproach to preventing these &quot;Sybil attacks&quot; is tohave a trusted agency certify identities. Thispaper shows that, without a logically centralizedauthority, Sybil...

Long short-term memory. Learning to store information over extended time intervals by recurrent backpropagation takes a very long time, mostly because of insufficient, decaying error backflow. We briefly review Hochreiter's (1991) analysis of this problem, then address it by introducing a novel, efficient, gradient-based method called long short-term memory (LSTM). Truncating the gradient where this does not do harm, LSTM can learn to bridge minimal time lags in excess of 1000 discrete-time steps by enforcing constant error flow through constant error carousels within special units. Multiplicative gate units learn to open and close access to the constant error flow. LSTM is local in space and time; its computational complexity per time step and weight is O(1). Our experiments with artificial data involve local, distributed, real-valued, and noisy pattern representations. In comparisons with real-time recurrent learning, back propagation through time, recurrent cascade correlation, Elman nets, and neural sequence chunking, LSTM leads to many more successful runs, and learns much faster. LSTM also solves complex, artificial long-time-lag tasks that have never been solved by previous recurrent network algorithms.

Toward Integrating Vehicular Clouds with IoT for Smart City Services Vehicular ad hoc networks, cloud computing, and the Internet of Things are among the emerging technology enablers offering a wide array of new application possibilities in smart urban spaces. These applications consist of smart building automation systems, healthcare monitoring systems, and intelligent and connected transportation, among others. The integration of IoT-based vehicular technologies will enrich services that are eventually going to ignite the proliferation of exciting and even more advanced technological marvels. However, depending on different requirements and design models for networking and architecture, such integration needs the development of newer communication architectures and frameworks. This work proposes a novel framework for architectural and communication design to effectively integrate vehicular networking clouds with IoT, referred to as VCoT, to materialize new applications that provision various IoT services through vehicular clouds. In this article, we particularly put emphasis on smart city applications deployed, operated, and controlled through LoRaWAN-based vehicular networks. LoraWAN, being a new technology, provides efficient and long-range communication possibilities. The article also discusses possible research issues in such an integration including data aggregation, security, privacy, data quality, and network coverage. These issues must be addressed in order to realize the VCoT paradigm deployment, and to provide insights for investors and key stakeholders in VCoT service provisioning. The article presents deep insights for different real-world application scenarios (i.e., smart homes, intelligent traffic light, and smart city) using VCoT for general control and automation along with their associated challenges. It also presents initial insights, through preliminary results, regarding data and resource management in IoT-based resource constrained environments through vehicular clouds.

Distributed multirobot localization In this paper, we present a new approach to the problem of simultaneously localizing a group of mobile robots capable of sensing one another. Each of the robots collects sensor data regarding its own motion and shares this information with the rest of the team during the update cycles. A single estimator, in the form of a Kalman filter, processes the available positioning information from all the members of the team and produces a pose estimate for every one of them. The equations for this centralized estimator can be written in a decentralized form, therefore allowing this single Kalman filter to be decomposed into a number of smaller communicating filters. Each of these filters processes the sensor data collected by its host robot. Exchange of information between the individual filters is necessary only when two robots detect each other and measure their relative pose. The resulting decentralized estimation schema, which we call collective localization, constitutes a unique means for fusing measurements collected from a variety of sensors with minimal communication and processing requirements. The distributed localization algorithm is applied to a group of three robots and the improvement in localization accuracy is presented. Finally, a comparison to the equivalent decentralized information filter is provided.

A simplified dual neural network for quadratic programming with its KWTA application. The design, analysis, and application of a new recurrent neural network for quadratic programming, called simplified dual neural network, are discussed. The analysis mainly concentrates on the convergence property and the computational complexity of the neural network. The simplified dual neural network is shown to be globally convergent to the exact optimal solution. The complexity of the neural network architecture is reduced with the number of neurons equal to the number of inequality constraints. Its application to k-winners-take-all (KWTA) operation is discussed to demonstrate how to solve problems with this neural network.

NETWRAP: An NDN Based Real-TimeWireless Recharging Framework for Wireless Sensor Networks Using vehicles equipped with wireless energy transmission technology to recharge sensor nodes over the air is a game-changer for traditional wireless sensor networks. The recharging policy regarding when to recharge which sensor nodes critically impacts the network performance. So far only a few works have studied such recharging policy for the case of using a single vehicle. In this paper, we propose NETWRAP, an N DN based Real Time Wireless Rech arging Protocol for dynamic wireless recharging in sensor networks. The real-time recharging framework supports single or multiple mobile vehicles. Employing multiple mobile vehicles provides more scalability and robustness. To efficiently deliver sensor energy status information to vehicles in real-time, we leverage concepts and mechanisms from named data networking (NDN) and design energy monitoring and reporting protocols. We derive theoretical results on the energy neutral condition and the minimum number of mobile vehicles required for perpetual network operations. Then we study how to minimize the total traveling cost of vehicles while guaranteeing all the sensor nodes can be recharged before their batteries deplete. We formulate the recharge optimization problem into a Multiple Traveling Salesman Problem with Deadlines (m-TSP with Deadlines), which is NP-hard. To accommodate the dynamic nature of node energy conditions with low overhead, we present an algorithm that selects the node with the minimum weighted sum of traveling time and residual lifetime. Our scheme not only improves network scalability but also ensures the perpetual operation of networks. Extensive simulation results demonstrate the effectiveness and efficiency of the proposed design. The results also validate the correctness of the theoretical analysis and show significant improvements that cut the number of nonfunctional nodes by half compared to the static scheme while maintaining the network overhead at the same level.

Finite-Time Adaptive Fuzzy Tracking Control Design for Nonlinear Systems. This paper addresses the finite-time tracking problem of nonlinear pure-feedback systems. Unlike the literature on traditional finite-time stabilization, in this paper the nonlinear system functions, including the bounding functions, are all totally unknown. Fuzzy logic systems are used to model those unknown functions. To present a finite-time control strategy, a criterion of semiglobal practical...

Convert Harm Into Benefit: A Coordination-Learning Based Dynamic Spectrum Anti-Jamming Approach This paper mainly investigates the multi-user anti-jamming spectrum access problem. Using the idea of “converting harm into benefit,” the malicious jamming signals projected by the enemy are utilized by the users as the coordination signals to guide spectrum coordination. An “internal coordination-external confrontation” multi-user anti-jamming access game model is constructed, and the existence of Nash equilibrium (NE) as well as correlated equilibrium (CE) is demonstrated. A coordination-learning based anti-jamming spectrum access algorithm (CLASA) is designed to achieve the CE of the game. Simulation results show the convergence, and effectiveness of the proposed CLASA algorithm, and indicate that our approach can help users confront the malicious jammer, and coordinate internal spectrum access simultaneously without information exchange. Last but not least, the fairness of the proposed approach under different jamming attack patterns is analyzed, which illustrates that this approach provides fair anti-jamming spectrum access opportunities under complicated jamming pattern.

Style Transfer from Non-Parallel Text by Cross-Alignment. This paper focuses on style transfer on the basis of non-parallel text. This is an instance of a broad family of problems including machine translation, decipherment, and sentiment modification. The key challenge is to separate the content from other aspects such as style. We assume a shared latent content distribution across different text corpora, and propose a method that leverages refined alignment of latent representations to perform style transfer. The transferred sentences from one style should match example sentences from the other style as a population. We demonstrate the effectiveness of this cross-alignment method on three tasks: sentiment modification, decipherment of word substitution ciphers, and recovery of word order.

An intelligent analyzer and understander of English The paper describes a working analysis and generation program for natural language, which handles paragraph length input. Its core is a system of preferential choice between deep semantic patterns, based on what we call “semantic density.” The system is contrasted:with syntax oriented linguistic approaches, and with theorem proving approaches to the understanding problem.

Hitting the right paraphrases in good time We present a random-walk-based approach to learning paraphrases from bilingual parallel corpora. The corpora are represented as a graph in which a node corresponds to a phrase, and an edge exists between two nodes if their corresponding phrases are aligned in a phrase table. We sample random walks to compute the average number of steps it takes to reach a ranking of paraphrases with better ones being "closer" to a phrase of interest. This approach allows "feature" nodes that represent domain knowledge to be built into the graph, and incorporates truncation techniques to prevent the graph from growing too large for efficiency. Current approaches, by contrast, implicitly presuppose the graph to be bipartite, are limited to finding paraphrases that are of length two away from a phrase, and do not generally permit easy incorporation of domain knowledge. Manual evaluation of generated output shows that our approach outperforms the state-of-the-art system of Callison-Burch (2008).

Re-examining machine translation metrics for paraphrase identification We propose to re-examine the hypothesis that automated metrics developed for MT evaluation can prove useful for paraphrase identification in light of the significant work on the development of new MT metrics over the last 4 years. We show that a meta-classifier trained using nothing but recent MT metrics outperforms all previous paraphrase identification approaches on the Microsoft Research Paraphrase corpus. In addition, we apply our system to a second corpus developed for the task of plagiarism detection and obtain extremely positive results. Finally, we conduct extensive error analysis and uncover the top systematic sources of error for a paraphrase identification approach relying solely on MT metrics. We release both the new dataset and the error analysis annotations for use by the community.

QANet: Combining Local Convolution with Global Self-Attention for Reading Comprehension. Current end-to-end machine reading and question answering (Qu0026A) models are primarily based on recurrent neural networks (RNNs) with attention. Despite their success, these models are often slow for both training and inference due to the sequential nature of RNNs. We propose a new Qu0026A model that does not require recurrent networks: It consists exclusively of attention and convolutions, yet achieves equivalent or better performance than existing models. On the SQuAD dataset, our model is 3x to 13x faster in training and 4x to 9x faster in inference. The speed-up gain allows us to train the model with much more data. We hence combine our model with data generated by backtranslation from a neural machine translation model. This data augmentation technique not only enhances the training examples but also diversifies the phrasing of the sentences, which results in immediate accuracy improvements. Our single model achieves 84.6 F1 score on the test set, which is significantly better than the best published F1 score of 81.8.

Integrating Transformer and Paraphrase Rules for Sentence Simplification. Sentence simplification aims to reduce the complexity of a sentence while retaining its original meaning. Current models for sentence simplification adopted ideas from ma- chine translation studies and implicitly learned simplification mapping rules from normal- simple sentence pairs. In this paper, we explore a novel model based on a multi-layer and multi-head attention architecture and we pro- pose two innovative approaches to integrate the Simple PPDB (A Paraphrase Database for Simplification), an external paraphrase knowledge base for simplification that covers a wide range of real-world simplification rules. The experiments show that the integration provides two major benefits: (1) the integrated model outperforms multiple state- of-the-art baseline models for sentence simplification in the literature (2) through analysis of the rule utilization, the model seeks to select more accurate simplification rules. The code and models used in the paper are available at this https URL Sanqiang/text_simplification.

Joint Copying and Restricted Generation for Paraphrase Many natural language generation tasks, such as abstractive summarization and text simplification, are paraphrase-orientated. In these tasks, copying and rewriting are two main writing modes. Most previous sequence-to-sequence (Seq2Seq) models use a single decoder and neglect this fact. In this paper, we develop a novel Seq2Seq model to fuse a copying decoder and a restricted generative decoder. The copying decoder finds the position to be copied based on a typical attention model. The generative decoder produces words limited in the source-specific vocabulary. To combine the two decoders and determine the final output, we develop a predictor to predict the mode of copying or rewriting. This predictor can be guided by the actual writing mode in the training data. We conduct extensive experiments on two different paraphrase datasets. The result shows that our model outperforms the state-of-the-art approaches in terms of both informativeness and language quality.

Microsoft Coco: Common Objects In Context We present a new dataset with the goal of advancing the state-of-the-art in object recognition by placing the question of object recognition in the context of the broader question of scene understanding. This is achieved by gathering images of complex everyday scenes containing common objects in their natural context. Objects are labeled using per-instance segmentations to aid in precise object localization. Our dataset contains photos of 91 objects types that would be easily recognizable by a 4 year old. With a total of 2.5 million labeled instances in 328k images, the creation of our dataset drew upon extensive crowd worker involvement via novel user interfaces for category detection, instance spotting and instance segmentation. We present a detailed statistical analysis of the dataset in comparison to PASCAL, ImageNet, and SUN. Finally, we provide baseline performance analysis for bounding box and segmentation detection results using a Deformable Parts Model.

No free lunch theorems for optimization A framework is developed to explore the connection between effective optimization algorithms and the problems they are solving. A number of “no free lunch” (NFL) theorems are presented which establish that for any algorithm, any elevated performance over one class of problems is offset by performance over another class. These theorems result in a geometric interpretation of what it means for an algorithm to be well suited to an optimization problem. Applications of the NFL theorems to information-theoretic aspects of optimization and benchmark measures of performance are also presented. Other issues addressed include time-varying optimization problems and a priori “head-to-head” minimax distinctions between optimization algorithms, distinctions that result despite the NFL theorems' enforcing of a type of uniformity over all algorithms

Response time in man-computer conversational transactions The literature concerning man-computer transactions abounds in controversy about the limits of "system response time" to a user's command or inquiry at a terminal. Two major semantic issues prohibit resolving this controversy. One issue centers around the question of "Response time to what?" The implication is that different human purposes and actions will have different acceptable or useful response times.

A Nonconservative LMI Condition for Stability of Switched Systems With Guaranteed Dwell Time. Ensuring stability of switched linear systems with a guaranteed dwell time is an important problem in control systems. Several methods have been proposed in the literature to address this problem, but unfortunately they provide sufficient conditions only. This technical note proposes the use of homogeneous polynomial Lyapunov functions in the non-restrictive case where all the subsystems are Hurwitz, showing that a sufficient condition can be provided in terms of an LMI feasibility test by exploiting a key representation of polynomials. Several properties are proved for this condition, in particular that it is also necessary for a sufficiently large degree of these functions. As a result, the proposed condition provides a sequence of upper bounds of the minimum dwell time that approximate it arbitrarily well. Some examples illustrate the proposed approach.

A recent survey of reversible watermarking techniques. The art of secretly hiding and communicating information has gained immense importance in the last two decades due to the advances in generation, storage, and communication technology of digital content. Watermarking is one of the promising solutions for tamper detection and protection of digital content. However, watermarking can cause damage to the sensitive information present in the cover work. Therefore, at the receiving end, the exact recovery of cover work may not be possible. Additionally, there exist certain applications that may not tolerate even small distortions in cover work prior to the downstream processing. In such applications, reversible watermarking instead of conventional watermarking is employed. Reversible watermarking of digital content allows full extraction of the watermark along with the complete restoration of the cover work. For the last few years, reversible watermarking techniques are gaining popularity because of its increasing applications in some important and sensitive areas, i.e., military communication, healthcare, and law-enforcement. Due to the rapid evolution of reversible watermarking techniques, a latest review of recent research in this field is highly desirable. In this survey, the performances of different reversible watermarking schemes are discussed on the basis of various characteristics of watermarking. However, the major focus of this survey is on prediction-error expansion based reversible watermarking techniques, whereby the secret information is hidden in the prediction domain through error expansion. Comparison of the different reversible watermarking techniques is provided in tabular form, and an analysis is carried out. Additionally, experimental comparison of some of the recent reversible watermarking techniques, both in terms of watermarking properties and computational time, is provided on a dataset of 300 images. Future directions are also provided for this potentially important field of watermarking.

Applications of Deep Reinforcement Learning in Communications and Networking: A Survey. This paper presents a comprehensive literature review on applications of deep reinforcement learning (DRL) in communications and networking. Modern networks, e.g., Internet of Things (IoT) and unmanned aerial vehicle (UAV) networks, become more decentralized and autonomous. In such networks, network entities need to make decisions locally to maximize the network performance under uncertainty of network environment. Reinforcement learning has been efficiently used to enable the network entities to obtain the optimal policy including, e.g., decisions or actions, given their states when the state and action spaces are small. However, in complex and large-scale networks, the state and action spaces are usually large, and the reinforcement learning may not be able to find the optimal policy in reasonable time. Therefore, DRL, a combination of reinforcement learning with deep learning, has been developed to overcome the shortcomings. In this survey, we first give a tutorial of DRL from fundamental concepts to advanced models. Then, we review DRL approaches proposed to address emerging issues in communications and networking. The issues include dynamic network access, data rate control, wireless caching, data offloading, network security, and connectivity preservation which are all important to next generation networks, such as 5G and beyond. Furthermore, we present applications of DRL for traffic routing, resource sharing, and data collection. Finally, we highlight important challenges, open issues, and future research directions of applying DRL.

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

Travel-Time Prediction of Bus Journey With Multiple Bus Trips Accurate travel-time prediction of public transport is essential for reliable journey planning in urban transportation systems. However, existing studies on bus travel-/arrival-time prediction often focus only on improving the prediction accuracy of a single bus trip. This is inadequate in modern public transportation systems, where a bus journey usually consists of multiple bus trips. In this paper, we investigate the problem of travel-time prediction for bus journeys that takes into account a passenger’s riding time on multiple bus trips, and also his/her waiting time at transfer points (interchange stations or bus stops). A novel framework is proposed to separately predict the riding and waiting time of a given journey from multiple datasets (i.e., historical bus trajectories, bus route, and road network), and combining the results to form the final travel-time prediction. We empirically determine the impact factors of bus riding times and develop a long short-term memory model that can accurately predict the riding time of each segment of the bus lines/routes. We also demonstrate that the waiting time at transfer points significantly impacts the total journey travel time, and estimating the waiting time is non-trivial as we cannot assume a fixed distribution waiting time. In order to accurately predict the waiting time, we introduce a novel interval-based historical average method that can efficiently address the correlation and sensitivity issues in waiting time prediction. Experiments on real-world data show that the proposed method notably outperforms six baseline approaches for all the scenarios considered.

Forecasting holiday daily tourist flow based on seasonal support vector regression with adaptive genetic algorithm. •The model of support vector regression with adaptive genetic algorithm and the seasonal mechanism is proposed.•Parameters selection and seasonal adjustment should be carefully selected.•We focus on latest and representative holiday daily data in China.•Two experiments are used to prove the effect of the model.•The AGASSVR is superior to AGA-SVR and BPNN.

Regression conformal prediction with random forests Regression conformal prediction produces prediction intervals that are valid, i.e., the probability of excluding the correct target value is bounded by a predefined confidence level. The most important criterion when comparing conformal regressors is efficiency; the prediction intervals should be as tight (informative) as possible. In this study, the use of random forests as the underlying model for regression conformal prediction is investigated and compared to existing state-of-the-art techniques, which are based on neural networks and k-nearest neighbors. In addition to their robust predictive performance, random forests allow for determining the size of the prediction intervals by using out-of-bag estimates instead of requiring a separate calibration set. An extensive empirical investigation, using 33 publicly available data sets, was undertaken to compare the use of random forests to existing state-of-the-art conformal predictors. The results show that the suggested approach, on almost all confidence levels and using both standard and normalized nonconformity functions, produced significantly more efficient conformal predictors than the existing alternatives.

Learning to Predict Bus Arrival Time From Heterogeneous Measurements via Recurrent Neural Network Bus arrival time prediction intends to improve the level of the services provided by transportation agencies. Intuitively, many stochastic factors affect the predictability of the arrival time, e.g., weather and local events. Moreover, the arrival time prediction for a current station is closely correlated with that of multiple passed stations. Motivated by the observations above, this paper propo...

Hybrid Spatio-Temporal Graph Convolutional Network: Improving Traffic Prediction with Navigation Data Traffic forecasting has recently attracted increasing interest due to the popularity of online navigation services, ridesharing and smart city projects. Owing to the non-stationary nature of road traffic, forecasting accuracy is fundamentally limited by the lack of contextual information. To address this issue, we propose the Hybrid Spatio-Temporal Graph Convolutional Network (H-STGCN), which is able to "deduce" future travel time by exploiting the data of upcoming traffic volume. Specifically, we propose an algorithm to acquire the upcoming traffic volume from an online navigation engine. Taking advantage of the piecewise-linear flow-density relationship, a novel transformer structure converts the upcoming volume into its equivalent in travel time. We combine this signal with the commonly-utilized travel-time signal, and then apply graph convolution to capture the spatial dependency. Particularly, we construct a compound adjacency matrix which reflects the innate traffic proximity. We conduct extensive experiments on real-world datasets. The results show that H-STGCN remarkably outperforms state-of-the-art methods in various metrics, especially for the prediction of non-recurring congestion.

Long-Term Traffic Speed Prediction Based on Multiscale Spatio-Temporal Feature Learning Network Speed plays a significant role in evaluating the evolution of traffic status, and predicting speed is one of the fundamental tasks for the intelligent transportation system. There exists a large number of works on speed forecast; however, the problem of long-term prediction for the next day is still not well addressed. In this paper, we propose a multiscale spatio-temporal feature learning network (MSTFLN) as the model to handle the challenging task of long-term traffic speed prediction for elevated highways. Raw traffic speed data collected from loop detectors every 5 min are transformed into spatial-temporal matrices; each matrix represents the one-day speed information, rows of the matrix indicate the numbers of loop detectors, and time intervals are denoted by columns. To predict the traffic speed of a certain day, nine speed matrices of three historical days with three different time scales are served as the input of MSTFLN. The proposed MSTFLN model consists of convolutional long short-term memories and convolutional neural networks. Experiments are evaluated using the data of three main elevated highways in Shanghai, China. The presented results demonstrate that our approach outperforms the state-of-the-art work and it can effectively predict the long-term speed information.

Transfer Knowledge between Cities The rapid urbanization has motivated extensive research on urban computing. It is critical for urban computing tasks to unlock the power of the diversity of data modalities generated by different sources in urban spaces, such as vehicles and humans. However, we are more likely to encounter the label scarcity problem and the data insufficiency problem when solving an urban computing task in a city where services and infrastructures are not ready or just built. In this paper, we propose a FLexible multimOdal tRAnsfer Learning (FLORAL) method to transfer knowledge from a city where there exist sufficient multimodal data and labels, to this kind of cities to fully alleviate the two problems. FLORAL learns semantically related dictionaries for multiple modalities from a source domain, and simultaneously transfers the dictionaries and labelled instances from the source into a target domain. We evaluate the proposed method with a case study of air quality prediction.

Space-time modeling of traffic flow. This paper discusses the application of space-time autoregressive integrated moving average (STARIMA) methodology for representing traffic flow patterns. Traffic flow data are in the form of spatial time series and are collected at specific locations at constant intervals of time. Important spatial characteristics of the space-time process are incorporated in the STARIMA model through the use of weighting matrices estimated on the basis of the distances among the various locations where data are collected. These matrices distinguish the space-time approach from the vector autoregressive moving average (VARMA) methodology and enable the model builders to control the number of the parameters that have to be estimated. The proposed models can be used for short-term forecasting of space-time stationary traffic-flow processes and for assessing the impact of traffic-flow changes on other parts of the network. The three-stage iterative space-time model building procedure is illustrated using 7.5min average traffic flow data for a set of 25 loop-detectors located at roads that direct to the centre of the city of Athens, Greece. Data for two months with different traffic-flow characteristics are modelled in order to determine the stability of the parameter estimation.

Model-based periodic event-triggered control for linear systems Periodic event-triggered control (PETC) is a control strategy that combines ideas from conventional periodic sampled-data control and event-triggered control. By communicating periodically sampled sensor and controller data only when needed to guarantee stability or performance properties, PETC is capable of reducing the number of transmissions significantly, while still retaining a satisfactory closed-loop behavior. In this paper, we will study observer-based controllers for linear systems and propose advanced event-triggering mechanisms (ETMs) that will reduce communication in both the sensor-to-controller channels and the controller-to-actuator channels. By exploiting model-based computations, the new classes of ETMs will outperform existing ETMs in the literature. To model and analyze the proposed classes of ETMs, we present two frameworks based on perturbed linear and piecewise linear systems, leading to conditions for global exponential stability and @?"2-gain performance of the resulting closed-loop systems in terms of linear matrix inequalities. The proposed analysis frameworks can be used to make tradeoffs between the network utilization on the one hand and the performance in terms of @?"2-gains on the other. In addition, we will show that the closed-loop performance realized by an observer-based controller, implemented in a conventional periodic time-triggered fashion, can be recovered arbitrarily closely by a PETC implementation. This provides a justification for emulation-based design. Next to centralized model-based ETMs, we will also provide a decentralized setup suitable for large-scale systems, where sensors and actuators are physically distributed over a wide area. The improvements realized by the proposed model-based ETMs will be demonstrated using numerical examples.

Affective social robots For human-robot interaction to proceed in a smooth, natural manner, robots must adhere to human social norms. One such human convention is the use of expressive moods and emotions as an integral part of social interaction. Such expressions are used to convey messages such as ''I'm happy to see you'' or ''I want to be comforted,'' and people's long-term relationships depend heavily on shared emotional experiences. Thus, we have developed an affective model for social robots. This generative model attempts to create natural, human-like affect and includes distinctions between immediate emotional responses, the overall mood of the robot, and long-term attitudes toward each visitor to the robot, with a focus on developing long-term human-robot relationships. This paper presents the general affect model as well as particular details of our implementation of the model on one robot, the Roboceptionist. In addition, we present findings from two studies that demonstrate the model's potential.

Rich Models for Steganalysis of Digital Images We describe a novel general strategy for building steganography detectors for digital images. The process starts with assembling a rich model of the noise component as a union of many diverse submodels formed by joint distributions of neighboring samples from quantized image noise residuals obtained using linear and nonlinear high-pass filters. In contrast to previous approaches, we make the model assembly a part of the training process driven by samples drawn from the corresponding cover- and stego-sources. Ensemble classifiers are used to assemble the model as well as the final steganalyzer due to their low computational complexity and ability to efficiently work with high-dimensional feature spaces and large training sets. We demonstrate the proposed framework on three steganographic algorithms designed to hide messages in images represented in the spatial domain: HUGO, edge-adaptive algorithm by Luo , and optimally coded ternary $\\pm {\\hbox{1}}$ embedding. For each algorithm, we apply a simple submodel-selection technique to increase the detection accuracy per model dimensionality and show how the detection saturates with increasing complexity of the rich model. By observing the differences between how different submodels engage in detection, an interesting interplay between the embedding and detection is revealed. Steganalysis built around rich image models combined with ensemble classifiers is a promising direction towards automatizing steganalysis for a wide spectrum of steganographic schemes.

Heterogeneous ensemble for feature drifts in data streams The nature of data streams requires classification algorithms to be real-time, efficient, and able to cope with high-dimensional data that are continuously arriving. It is a known fact that in high-dimensional datasets, not all features are critical for training a classifier. To improve the performance of data stream classification, we propose an algorithm called HEFT-Stream (H eterogeneous E nsemble with F eature drifT for Data Streams ) that incorporates feature selection into a heterogeneous ensemble to adapt to different types of concept drifts. As an example of the proposed framework, we first modify the FCBF [13] algorithm so that it dynamically update the relevant feature subsets for data streams. Next, a heterogeneous ensemble is constructed based on different online classifiers, including Online Naive Bayes and CVFDT [5]. Empirical results show that our ensemble classifier outperforms state-of-the-art ensemble classifiers (AWE [15] and OnlineBagging [21]) in terms of accuracy, speed, and scalability. The success of HEFT-Stream opens new research directions in understanding the relationship between feature selection techniques and ensemble learning to achieve better classification performance.

Orientation-aware RFID tracking with centimeter-level accuracy. RFID tracking attracts a lot of research efforts in recent years. Most of the existing approaches, however, adopt an orientation-oblivious model. When tracking a target whose orientation changes, those approaches suffer from serious accuracy degradation. In order to achieve target tracking with pervasive applicability in various scenarios, we in this paper propose OmniTrack, an orientation-aware RFID tracking approach. Our study discovers the linear relationship between the tag orientation and the phase change of the backscattered signals. Based on this finding, we propose an orientation-aware phase model to explicitly quantify the respective impact of the read-tag distance and the tag's orientation. OmniTrack addresses practical challenges in tracking the location and orientation of a mobile tag. Our experimental results demonstrate that OmniTrack achieves centimeter-level location accuracy and has significant advantages in tracking targets with varing orientations, compared to the state-of-the-art approaches.

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

Federated Learning via Intelligent Reflecting Surface Over-the-air computation (AirComp) based federated learning (FL) is capable of achieving fast model aggregation by exploiting the waveform superposition property of multiple-access channels. However, the model aggregation performance is severely limited by the unfavorable wireless propagation channels. In this paper, we propose to leverage intelligent reflecting surface (IRS) to achieve fast yet reliable model aggregation for AirComp-based FL. To optimize the learning performance, we present the convergence analysis of our proposed IRS-assisted AirComp-based FL system, based on which we propose to maximize the number of scheduled devices of each communication round under certain mean-squared error (MSE) requirements. To tackle the formulated highly-intractable problem, we propose a two-step optimization framework. Specifically, we induce the sparsity of device selection in the first step, followed by solving a series of MSE minimization problems to find the maximum feasible device set in the second step. We then propose an alternating optimization framework, supported by the difference-of-convex programming for low-rank optimization, to efficiently design the aggregation beamformers at the BS and phase shifts at the IRS. Simulation results demonstrate that our proposed algorithm and the deployment of an IRS can achieve a higher FL prediction accuracy than the baseline schemes.

Relay-Assisted Cooperative Federated Learning Federated learning (FL) has recently emerged as a promising technology to enable artificial intelligence (AI) at the network edge, where distributed mobile devices collaboratively train a shared AI model under the coordination of an edge server. To significantly improve the communication efficiency of FL, over-the-air computation allows a large number of mobile devices to concurrently upload their local models by exploiting the superposition property of wireless multi-access channels. Due to wireless channel fading, the model aggregation error at the edge server is dominated by the weakest channel among all devices, causing severe straggler issues. In this paper, we propose a relay-assisted cooperative FL scheme to effectively address the straggler issue. In particular, we deploy multiple half-duplex relays to cooperatively assist the devices in uploading the local model updates to the edge server. The nature of the over-the-air computation poses system objectives and constraints that are distinct from those in traditional relay communication systems. Moreover, the strong coupling between the design variables renders the optimization of such a system challenging. To tackle the issue, we propose an alternating-optimization-based algorithm to optimize the transceiver and relay operation with low complexity. Then, we analyze the model aggregation error in a single-relay case and show that our relay-assisted scheme achieves a smaller error than the one without relays provided that the relay transmit power and the relay channel gains are sufficiently large. The analysis provides critical insights on relay deployment in the implementation of cooperative FL. Extensive numerical results show that our design achieves faster convergence compared with state-of-the-art schemes.

Delay Analysis of Wireless Federated Learning Based on Saddle Point Approximation and Large Deviation Theory Federated learning (FL) is a collaborative machine learning paradigm, which enables deep learning model training over a large volume of decentralized data residing in mobile devices without accessing clients’ private data. Driven by the ever increasing demand for model training of mobile applications or devices, a vast majority of FL tasks are implemented over wireless fading channels. Due to the ...

A Compressive Sensing Approach for Federated Learning Over Massive MIMO Communication Systems Federated learning is a privacy-preserving approach to train a global model at a central server by collaborating with wireless devices, each with its own local training data set. In this paper, we present a compressive sensing approach for federated learning over massive multiple-input multiple-output communication systems in which the central server equipped with a massive antenna array communicates with the wireless devices. One major challenge in system design is to reconstruct local gradient vectors accurately at the central server, which are computed-and-sent from the wireless devices. To overcome this challenge, we first establish a transmission strategy to construct sparse transmitted signals from the local gradient vectors at the devices. We then propose a compressive sensing algorithm enabling the server to iteratively find the linear minimum-mean-square-error (LMMSE) estimate of the transmitted signal by exploiting its sparsity. We also derive an analytical threshold for the residual error at each iteration, to design the stopping criterion of the proposed algorithm. We show that for a sparse transmitted signal, the proposed algorithm requires less computationally complexity than LMMSE. Simulation results demonstrate that the presented approach outperforms conventional linear beamforming approaches and reduces the performance gap between federated learning and centralized learning with perfect reconstruction.

Cell-Free Massive MIMO for Wireless Federated Learning This paper proposes a novel scheme for cell-free massive multiple-input multiple-output (CFmMIMO) networks to support any federated learning (FL) framework. This scheme allows each instead of all the iterations of the FL framework to happen in a large-scale coherence time to guarantee a stable operation of an FL process. To show how to optimize the FL performance using this proposed scheme, we consider an existing FL framework as an example and target FL training time minimization for this framework. An optimization problem is then formulated to jointly optimize the local accuracy, transmit power, data rate, and users' processing frequency. This mixed-timescale stochastic nonconvex problem captures the complex interactions among the training time, and transmission and computation of training updates of one FL process. By employing the online successive convex approximation approach, we develop a new algorithm to solve the formulated problem with proven convergence to the neighbourhood of its stationary points. Our numerical results confirm that the presented joint design reduces the training time by up to 55% over baseline approaches. They also show that CFmMIMO here requires the lowest training time for FL processes compared with cell-free time-division multiple access massive MIMO and collocated massive MIMO.

Federated Learning: Challenges, Methods, and Future Directions Federated learning involves training statistical models over remote devices or siloed data centers, such as mobile phones or hospitals, while keeping data localized. Training in heterogeneous and potentially massive networks introduces novel challenges that require a fundamental departure from standard approaches for large-scale machine learning, distributed optimization, and privacy-preserving data analysis. In this article, we discuss the unique characteristics and challenges of federated learning, provide a broad overview of current approaches, and outline several directions of future work that are relevant to a wide range of research communities.

Fuzzy logic in control systems: fuzzy logic controller. I.

An introduction to ROC analysis Receiver operating characteristics (ROC) graphs are useful for organizing classifiers and visualizing their performance. ROC graphs are commonly used in medical decision making, and in recent years have been used increasingly in machine learning and data mining research. Although ROC graphs are apparently simple, there are some common misconceptions and pitfalls when using them in practice. The purpose of this article is to serve as an introduction to ROC graphs and as a guide for using them in research.

Markov games as a framework for multi-agent reinforcement learning In the Markov decision process (MDP) formalization of reinforcement learning, a single adaptive agent interacts with an environment defined by a probabilistic transition function. In this solipsis-tic view, secondary agents can only be part of the environment and are therefore fixed in their behavior. The framework of Markov games allows us to widen this view to include multiple adaptive agents with interacting or competing goals. This paper considers a step in this direction in which exactly two agents with diametrically opposed goals share an environment. It describes a Q-learning-like algorithm for finding optimal policies and demonstrates its application to a simple two-player game in which the optimal policy is probabilistic.

Scalable and efficient provable data possession. Storage outsourcing is a rising trend which prompts a number of interesting security issues, many of which have been extensively investigated in the past. However, Provable Data Possession (PDP) is a topic that has only recently appeared in the research literature. The main issue is how to frequently, efficiently and securely verify that a storage server is faithfully storing its client's (potentially very large) outsourced data. The storage server is assumed to be untrusted in terms of both security and reliability. (In other words, it might maliciously or accidentally erase hosted data; it might also relegate it to slow or off-line storage.) The problem is exacerbated by the client being a small computing device with limited resources. Prior work has addressed this problem using either public key cryptography or requiring the client to outsource its data in encrypted form. In this paper, we construct a highly efficient and provably secure PDP technique based entirely on symmetric key cryptography, while not requiring any bulk encryption. Also, in contrast with its predecessors, our PDP technique allows outsourcing of dynamic data, i.e, it efficiently supports operations, such as block modification, deletion and append.

Cognitive Cars: A New Frontier for ADAS Research This paper provides a survey of recent works on cognitive cars with a focus on driver-oriented intelligent vehicle motion control. The main objective here is to clarify the goals and guidelines for future development in the area of advanced driver-assistance systems (ADASs). Two major research directions are investigated and discussed in detail: 1) stimuli–decisions–actions, which focuses on the driver side, and 2) perception enhancement–action-suggestion–function-delegation, which emphasizes the ADAS side. This paper addresses the important achievements and major difficulties of each direction and discusses how to combine the two directions into a single integrated system to obtain safety and comfort while driving. Other related topics, including driver training and infrastructure design, are also studied.

Online Prediction of Driver Distraction Based on Brain Activity Patterns This paper presents a new computational framework for early detection of driver distractions (map viewing) using brain activity measured by electroencephalographic (EEG) signals. Compared with most studies in the literature, which are mainly focused on the classification of distracted and nondistracted periods, this study proposes a new framework to prospectively predict the start and end of a distraction period, defined by map viewing. The proposed prediction algorithm was tested on a data set of continuous EEG signals recorded from 24 subjects. During the EEG recordings, the subjects were asked to drive from an initial position to a destination using a city map in a simulated driving environment. The overall accuracy values for the prediction of the start and the end of map viewing were 81% and 70%, respectively. The experimental results demonstrated that the proposed algorithm can predict the start and end of map viewing with relatively high accuracy and can be generalized to individual subjects. The outcome of this study has a high potential to improve the design of future intelligent navigation systems. Prediction of the start of map viewing can be used to provide route information based on a driver's needs and consequently avoid map-viewing activities. Prediction of the end of map viewing can be used to provide warnings for potential long map-viewing durations. Further development of the proposed framework and its applications in driver-distraction predictions are also discussed.

Adaptive fuzzy tracking control for switched uncertain strict-feedback nonlinear systems. •Adaptive tracking control for switched strict-feedback nonlinear systems is proposed.•The generalized fuzzy hyperbolic model is used to approximate nonlinear functions.•The designed controller has fewer design parameters comparing with existing methods.

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

Deep Residual Learning for Image Recognition Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously. We explicitly reformulate the layers as learning residual functions with reference to the layer inputs, instead of learning unreferenced functions. We provide comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth. On the ImageNet dataset we evaluate residual nets with a depth of up to 152 layers - 8× deeper than VGG nets [40] but still having lower complexity. An ensemble of these residual nets achieves 3.57% error on the ImageNet test set. This result won the 1st place on the ILSVRC 2015 classification task. We also present analysis on CIFAR-10 with 100 and 1000 layers. The depth of representations is of central importance for many visual recognition tasks. Solely due to our extremely deep representations, we obtain a 28% relative improvement on the COCO object detection dataset. Deep residual nets are foundations of our submissions to ILSVRC & COCO 2015 competitions1, where we also won the 1st places on the tasks of ImageNet detection, ImageNet localization, COCO detection, and COCO segmentation.

Horizontal Pyramid Matching for Person Re-Identification Despite the remarkable progress in person re-identification (Re-ID), such approaches still suffer from the failure cases where the discriminative body parts are missing. To mitigate this type of failure, we propose a simple yet effective Horizontal Pyramid Matching (HPM) approach to fully exploit various partial information of a given person, so that correct person candidates can be identified even if some key parts are missing. With HPM, we make the following contributions to produce more robust feature representations for the Re-ID task: 1) we learn to classify using partial feature representations at different horizontal pyramid scales, which successfully enhance the discriminative capabilities of various person parts; 2) we exploit average and max pooling strategies to account for person-specific discriminative information in a global-local manner. To validate the effectiveness of our proposed HPM method, extensive experiments are conducted on three popular datasets including Market-1501, DukeMTMC-ReID and CUHK03. Respectively, we achieve mAP scores of 83.1%, 74.5% and 59.7% on these challenging benchmarks, which are the new state-of-the-arts.

FD-GAN: Pose-guided Feature Distilling GAN for Robust Person Re-identification. Person re-identification (reID) is an important task that requires to retrieve a person's images from an image dataset, given one image of the person of interest. For learning robust person features, the pose variation of person images is one of the key challenges. Existing works targeting the problem either perform human alignment, or learn human-region-based representations. Extra pose information and computational cost is generally required for inference. To solve this issue, a Feature Distilling Generative Adversarial Network (FD-GAN) is proposed for learning identity-related and pose-unrelated representations. It is a novel framework based on a Siamese structure with multiple novel discriminators on human poses and identities. In addition to the discriminators, a novel same-pose loss is also integrated, which requires appearance of a same person's generated images to be similar. After learning pose-unrelated person features with pose guidance, no auxiliary pose information and additional computational cost is required during testing. Our proposed FD-GAN achieves state-of-the-art performance on three person reID datasets, which demonstrates that the effectiveness and robust feature distilling capability of the proposed FD-GAN.(double dagger double dagger)

Progressive Attention Guided Recurrent Network for Salient Object Detection Effective convolutional features play an important role in saliency estimation but how to learn powerful features for saliency is still a challenging task. FCN-based methods directly apply multi-level convolutional features without distinction, which leads to sub-optimal results due to the distraction from redundant details. In this paper, we propose a novel attention guided network which selectively integrates multi-level contextual information in a progressive manner. Attentive features generated by our network can alleviate distraction of background thus achieve better performance. On the other hand, it is observed that most of existing algorithms conduct salient object detection by exploiting side-output features of the backbone feature extraction network. However, shallower layers of backbone network lack the ability to obtain global semantic information, which limits the effective feature learning. To address the problem, we introduce multi-path recurrent feedback to enhance our proposed progressive attention driven framework. Through multi-path recurrent connections, global semantic information from the top convolutional layer is transferred to shallower layers, which intrinsically refines the entire network. Experimental results on six benchmark datasets demonstrate that our algorithm performs favorably against the state-of-the-art approaches.

CamStyle: A Novel Data Augmentation Method for Person Re-identification. Person re-identification (re-ID) is a cross-camera retrieval task that suffers from image style variations caused by different cameras. The art implicitly addresses this problem by learning a camera-invariant descriptor subspace. In this paper, we explicitly consider this challenge by introducing camera style (CamStyle). CamStyle can serve as a data augmentation approach that reduces the risk of d...

Progressive Growing of GANs for Improved Quality, Stability, and Variation. We describe a new training methodology for generative adversarial networks. The key idea is to grow both the generator and discriminator progressively: starting from a low resolution, we add new layers that model increasingly fine details as training progresses. This both speeds the training up and greatly stabilizes it, allowing us to produce images of unprecedented quality, e.g., CelebA images at 1024^2. We also propose a simple way to increase the variation in generated images, and achieve a record inception score of 8.80 in unsupervised CIFAR10. Additionally, we describe several implementation details that are important for discouraging unhealthy competition between the generator and discriminator. Finally, we suggest a new metric for evaluating GAN results, both in terms of image quality and variation. As an additional contribution, we construct a higher-quality version of the CelebA dataset.

Dual Attention Network For Scene Segmentation In this paper, we address the scene segmentation task by capturing rich contextual dependencies based on the self-attention mechanism. Unlike previous works that capture contexts by multi-scale feature fusion, we propose a Dual Attention Network (DANet) to adaptively integrate local features with their global dependencies. Specifically, we append two types of attention modules on top of dilated FCN, which model the semantic interdependencies in spatial and channel dimensions respectively. The position attention module selectively aggregates the feature at each position by a weighted sum of the features at all positions. Similar features would be related to each other regardless of their distances. Meanwhile, the channel attention module selectively emphasizes interdependent channel maps by integrating associated features among all channel maps. We sum the outputs of the two attention modules to further improve feature representation which contributes to more precise segmentation results. We achieve new state-of-the-art segmentation performance on three challenging scene segmentation datasets, i.e., Cityscapes, PASCAL Context and COCO Stuff dataset. In particular, a Mean IoU score of 81.5% on Cityscapes test set is achieved without using coarse data.(1).

IMD2020: A Large-Scale Annotated Dataset Tailored for Detecting Manipulated Images Witnessing impressive results of deep nets in a number of computer vision problems, the image forensic community has begun to utilize them in the challenging domain of detecting manipulated visual content. One of the obstacles to replicate the success of deep nets here is absence of diverse datasets tailored for training and testing of image forensic methods. Such datasets need to be designed to capture wide and complex types of systematic noise and intrinsic artifacts of images in order to avoid overfitting of learning methods to just a narrow set of camera types or types of manipulations. These artifacts are brought into visual content by various components of the image acquisition process as well as the manipulating process.In this paper, we introduce two novel datasets. First, we identified the majority of camera brands and models on the market, which resulted in 2,322 camera models. Then, we collected a dataset of 35,000 real images captured by these camera models. Moreover, we also created the same number of digitally manipulated images by using a large variety of core image manipulation methods as well we advanced ones such as GAN or Inpainting resulting in a dataset of 70,000 images. In addition to this dataset, we also created a dataset of 2,000 "real-life" (uncontrolled) manipulated images. They are made by unknown people and downloaded from Internet. The real versions of these images also have been found and are provided. We also manually created binary masks localizing the exact manipulated areas of these images. Both datasets are publicly available for the research community at http://staff.utia.cas.cz/novozada/db.

Open Set Domain Adaptation When the training and the test data belong to different domains, the accuracy of an object classifier is significantly reduced. Therefore, several algorithms have been proposed in the last years to diminish the so called domain shift between datasets. However, all available evaluation protocols for domain adaptation describe a closed set recognition task, where both domains, namely source and target, contain exactly the same object classes. In this work, we also explore the field of domain adaptation in open sets, which is a more realistic scenario where only a few categories of interest are shared between source and target data. Therefore, we propose a method that fits in both closed and open set scenarios. The approach learns a mapping from the source to the target domain by jointly solving an assignment problem that labels those target instances that potentially belong to the categories of interest present in the source dataset. A thorough evaluation shows that our approach outperforms the state-of-the-art.

Person Re-Identification With Metric Learning Using Privileged Information. Despite the promising progress made in recent years, person re-identification remains a challenging task due to complex variations in human appearances from different camera views. This paper presents a logistic discriminant metric learning method for this challenging problem. Different with most existing metric learning algorithms, it exploits both original data and auxiliary data during training...

Citywide traffic congestion estimation with social media. Conventional traffic congestion estimation approaches require the deployment of traffic sensors or large-scale probe vehicles. The high cost of deploying and maintaining these equipments largely limits their spatial-temporal coverage. This paper proposes an alternative solution with lower cost and wider spatial coverage by exploring traffic related information from Twitter. By regarding each Twitter user as a traffic monitoring sensor, various real-time traffic information can be collected freely from each corner of the city. However, there are two major challenges for this problem. Firstly, the congestion related information extracted directly from real-time tweets are very sparse due both to the low resolution of geographic location mentioned in the tweets and the inherent sparsity nature of Twitter data. Secondly, the traffic event information coming from Twitter can be multi-typed including congestion, accident, road construction, etc. It is non-trivial to model the potential impacts of diverse traffic events on traffic congestion. We propose to enrich the sparse real-time tweets from two directions: 1) mining the spatial and temporal correlations of the road segments in congestion from historical data, and 2) applying auxiliary information including social events and road features for help. We finally propose a coupled matrix and tensor factorization model to effectively integrate rich information for Citywide Traffic Congestion Eestimation (CTCE). Extensive evaluations on Twitter data and 500 million public passenger buses GPS data on nearly 700 mile roads of Chicago demonstrate the efficiency and effectiveness of the proposed approach.

Social navigation: techniques for building more usable systems

Hybrid local and global descriptor enhanced with colour information. Feature extraction is one of the most important steps in computer vision tasks such as object recognition, image retrieval and image classification. It describes an image by a set of descriptors where the best one gives a high quality description and a low computation. In this study, the authors propose a novel descriptor called histogram of local and global features using speeded up robust featur...

A Cyclic Game for Service-Oriented Resource Allocation in Edge Computing Existing works adopt the Edge-Oriented Resource Allocation (EORA) scheme, in which edge nodes cache services and schedule user requests to distribute workloads over cloud and edge nodes, so as to achieve high-quality services and low latency. Unfortunately, EORA does not fully take into account the fact that service providers are sometimes independent from the edge operators with their own objectives. To deal with the conflict and cooperation between service providers and edge nodes, we devise a service-oriented resource allocation (SORA) scheme, where edge nodes and service providers adjust their resource allocations to provide requested services. We first prove that such resource allocation problem is NP-hard. We then propose a three-sided cyclic game (3CG) involving users, edge nodes, and service providers who make their individual decisions by choosing respectively high-quality services, high-value users, and cost-effective edge nodes for service deployment. Based on 3CG, we prove the existence and approximation ratio of pure-strategy Nash equilibriums (NEs). We also develop both centralized and distributed approximate algorithms for resource allocation. Finally, extensive experimental results validate the effectiveness and convergence of the proposed algorithms.

Episodic Training For Domain Generalization Domain generalization (DG) is the challenging and topical problem of learning models that generalize to novel testing domains with different statistics than a set of known training domains. The simple approach of aggregating data from all source domains and training a single deep neural network end-to-end on all the data provides a surprisingly strong baseline that surpasses many prior published methods. In this paper we build on this strong baseline by designing an episodic training procedure that trains a single deep network in a way that exposes it to the domain shift that characterises a novel domain at runtime. Specifically, we decompose a deep network into feature extractor and classifier components, and then train each component by simulating it interacting with a partner who is badly tuned for the current domain. This makes both components more robust, ultimately leading to our networks producing state-of-the-art performance on three DG benchmarks. Furthermore, we consider the pervasive workflow of using an ImageNet trained CNN as a fixed feature extractor for downstream recognition tasks. Using the Visual Decathlon benchmark, we demonstrate that our episodic-DG training improves the performance of such a general purpose feature extractor by explicitly training a feature for robustness to novel problems. This shows that DG training can benefit standard practice in computer vision.

Scatter Component Analysis: A Unified Framework for Domain Adaptation and Domain Generalization. This paper addresses classification tasks on a particular target domain in which labeled training data are only available from source domains different from (but related to) the target. Two closely related frameworks, domain adaptation and domain generalization, are concerned with such tasks, where the only difference between those frameworks is the availability of the unlabeled target data: domai...

Diversity in Machine Learning. Machine learning methods have achieved good performance and been widely applied in various real-world applications. They can learn the model adaptively and be better fit for special requirements of different tasks. Generally, a good machine learning system is composed of plentiful training data, a good model training process, and an accurate inference. Many factors can affect the performance of the machine learning process, among which the diversity of the machine learning process is an important one. The diversity can help each procedure to guarantee a totally good machine learning: diversity of the training data ensures that the training data can provide more discriminative information for the model, diversity of the learned model (diversity in parameters of each model or diversity among different base models) makes each parameter/model capture unique or complement information and the diversity in inference can provide multiple choices each of which corresponds to a specific plausible local optimal result. Even though diversity plays an important role in the machine learning process, there is no systematical analysis of the diversification in the machine learning system. In this paper, we systematically summarize the methods to make data diversification, model diversification, and inference diversification in the machine learning process. In addition, the typical applications where the diversity technology improved the machine learning performance have been surveyed including the remote sensing imaging tasks, machine translation, camera relocalization, image segmentation, object detection, topic modeling, and others. Finally, we discuss some challenges of the diversity technology in machine learning and point out some directions in future work. Our analysis provides a deeper understanding of the diversity technology in machine learning tasks and hence can help design and learn more effective models for real-world applications.

Open Set Domain Adaptation When the training and the test data belong to different domains, the accuracy of an object classifier is significantly reduced. Therefore, several algorithms have been proposed in the last years to diminish the so called domain shift between datasets. However, all available evaluation protocols for domain adaptation describe a closed set recognition task, where both domains, namely source and target, contain exactly the same object classes. In this work, we also explore the field of domain adaptation in open sets, which is a more realistic scenario where only a few categories of interest are shared between source and target data. Therefore, we propose a method that fits in both closed and open set scenarios. The approach learns a mapping from the source to the target domain by jointly solving an assignment problem that labels those target instances that potentially belong to the categories of interest present in the source dataset. A thorough evaluation shows that our approach outperforms the state-of-the-art.

Person re-identification by descriptive and discriminative classification Person re-identification, i.e., recognizing a single person across spatially disjoint cameras, is an important task in visual surveillance. Existing approaches either try to find a suitable description of the appearance or learn a discriminative model. Since these different representational strategies capture a large extent of complementary information we propose to combine both approaches. First, given a specific query, we rank all samples according to a feature-based similarity, where appearance is modeled by a set of region covariance descriptors. Next, a discriminative model is learned using boosting for feature selection, which provides a more specific classifier. The proposed approach is demonstrated on two datasets, where we show that the combination of a generic descriptive statistical model and a discriminatively learned feature-based model attains considerably better results than the individual models alone. In addition, we give a comparison to the state-of-the-art on a publicly available benchmark dataset.

Scalable Person Re-Identification: A Benchmark This paper contributes a new high quality dataset for person re-identification, named \"Market-1501\". Generally, current datasets: 1) are limited in scale, 2) consist of hand-drawn bboxes, which are unavailable under realistic settings, 3) have only one ground truth and one query image for each identity (close environment). To tackle these problems, the proposed Market-1501 dataset is featured in three aspects. First, it contains over 32,000 annotated bboxes, plus a distractor set of over 500K images, making it the largest person re-id dataset to date. Second, images in Market-1501 dataset are produced using the Deformable Part Model (DPM) as pedestrian detector. Third, our dataset is collected in an open system, where each identity has multiple images under each camera. As a minor contribution, inspired by recent advances in large-scale image search, this paper proposes an unsupervised Bag-of-Words descriptor. We view person re-identification as a special task of image search. In experiment, we show that the proposed descriptor yields competitive accuracy on VIPeR, CUHK03, and Market-1501 datasets, and is scalable on the large-scale 500k dataset.

Accurate Self-Localization in RFID Tag Information Grids Using FIR Filtering Grid navigation spaces nested with the radio-frequency identification (RFID) tags are promising for industrial and other needs, because each tag can deliver information about a local two-dimensional or three-dimensional surrounding. The approach, however, requires high accuracy in vehicle self-localization. Otherwise, errors may lead to collisions; possibly even fatal. We propose a new extended finite impulse response (EFIR) filtering algorithm and show that it meets this need. The EFIR filter requires an optimal averaging interval, but does not involve the noise statistics which are often not well known to the engineer. It is more accurate than the extended Kalman filter (EKF) under real operation conditions and its iterative algorithm has the Kalman form. Better performance of the proposed EFIR filter is demonstrated based on extensive simulations in a comparison to EKF, which is widely used in RFID tag grids. We also show that errors in noise covariances may provoke divergence in EKF, whereas the EFIR filter remains stable and is thus more robust.

Distributed Representations, Simple Recurrent Networks, And Grammatical Structure In this paper three problems for a connectionist account of language are considered:1. What is the nature of linguistic representations?2. How can complex structural relationships such as constituent structure be represented?3. How can the apparently open-ended nature of language be accommodated by a fixed-resource system?Using a prediction task, a simple recurrent network (SRN) is trained on multiclausal sentences which contain multiply-embedded relative clauses. Principal component analysis of the hidden unit activation patterns reveals that the network solves the task by developing complex distributed representations which encode the relevant grammatical relations and hierarchical constituent structure. Differences between the SRN state representations and the more traditional pushdown store are discussed in the final section.

Social navigation: techniques for building more usable systems

A study on the use of non-parametric tests for analyzing the evolutionary algorithms' behaviour: a case study on the CEC'2005 Special Session on Real Parameter Optimization In recent years, there has been a growing interest for the experimental analysis in the field of evolutionary algorithms. It is noticeable due to the existence of numerous papers which analyze and propose different types of problems, such as the basis for experimental comparisons of algorithms, proposals of different methodologies in comparison or proposals of use of different statistical techniques in algorithms’ comparison.In this paper, we focus our study on the use of statistical techniques in the analysis of evolutionary algorithms’ behaviour over optimization problems. A study about the required conditions for statistical analysis of the results is presented by using some models of evolutionary algorithms for real-coding optimization. This study is conducted in two ways: single-problem analysis and multiple-problem analysis. The results obtained state that a parametric statistical analysis could not be appropriate specially when we deal with multiple-problem results. In multiple-problem analysis, we propose the use of non-parametric statistical tests given that they are less restrictive than parametric ones and they can be used over small size samples of results. As a case study, we analyze the published results for the algorithms presented in the CEC’2005 Special Session on Real Parameter Optimization by using non-parametric test procedures.

Labels and event processes in the Asbestos operating system Asbestos, a new operating system, provides novel labeling and isolation mechanisms that help contain the effects of exploitable software flaws. Applications can express a wide range of policies with Asbestos's kernel-enforced labels, including controls on interprocess communication and system-wide information flow. A new event process abstraction defines lightweight, isolated contexts within a single process, allowing one process to act on behalf of multiple users while preventing it from leaking any single user's data to others. A Web server demonstration application uses these primitives to isolate private user data. Since the untrusted workers that respond to client requests are constrained by labels, exploited workers cannot directly expose user data except as allowed by application policy. The server application requires 1.4 memory pages per user for up to 145,000 users and achieves connection rates similar to Apache, demonstrating that additional security can come at an acceptable cost.

Adaptive Consensus Control for a Class of Nonlinear Multiagent Time-Delay Systems Using Neural Networks Because of the complicity of consensus control of nonlinear multiagent systems in state time-delay, most of previous works focused only on linear systems with input time-delay. An adaptive neural network (NN) consensus control method for a class of nonlinear multiagent systems with state time-delay is proposed in this paper. The approximation property of radial basis function neural networks (RBFNNs) is used to neutralize the uncertain nonlinear dynamics in agents. An appropriate Lyapunov-Krasovskii functional, which is obtained from the derivative of an appropriate Lyapunov function, is used to compensate the uncertainties of unknown time delays. It is proved that our proposed approach guarantees the convergence on the basis of Lyapunov stability theory. The simulation results of a nonlinear multiagent time-delay system and a multiple collaborative manipulators system show the effectiveness of the proposed consensus control algorithm.

Robust Sparse Linear Discriminant Analysis Linear discriminant analysis (LDA) is a very popular supervised feature extraction method and has been extended to different variants. However, classical LDA has the following problems: 1) The obtained discriminant projection does not have good interpretability for features. 2) LDA is sensitive to noise. 3) LDA is sensitive to the selection of number of projection directions. In this paper, a novel feature extraction method called robust sparse linear discriminant analysis (RSLDA) is proposed to solve the above problems. Specifically, RSLDA adaptively selects the most discriminative features for discriminant analysis by introducing the l2;1 norm. An orthogonal matrix and a sparse matrix are also simultaneously introduced to guarantee that the extracted features can hold the main energy of the original data and enhance the robustness to noise, and thus RSLDA has the potential to perform better than other discriminant methods. Extensive experiments on six databases demonstrate that the proposed method achieves the competitive performance compared with other state-of-the-art feature extraction methods. Moreover, the proposed method is robust to the noisy data. IEEE

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

Adaptive Interference Avoidance for Dynamic Wireless Systems: A Game Theoretic Approach In this paper, we present an adaptive algorithm for interference avoidance, which can be applied in a distributed manner by active users in a CDMA wireless system to obtain optimal codewords and powers for specified target signal-to-interference plus noise ratio (SINR). The algorithm is derived using a game-theoretic approach in which separable cost functions with respect to codeword and power are...

Optimal distributed interference avoidance: potential game and learning. This article studies the problem of distributed interference avoidance (IA) through channel selection for distributed wireless networks, where mutual interference only occurs among nearby users. First, an interference graph is used to characterise the limited range of interference, and then the distributed IA problem is formulated as a graph colouring problem. Because solving the graph colouring problem is non-deterministic polynomial hard even in a centralised manner, the task of obtaining the optimal channel selection profile distributively is challenging. We formulate this problem as a channel selection game, which is proved to be an exact potential game with the weighted aggregate interference serving as the potential function. On the basis of this, a distributed learning algorithm is proposed to achieve the optimal channel selection profile that constitutes an optimal Nash equilibrium point of the channel selection game. The proposed learning algorithm is fully distributed because it needs information about neither the network topology nor the actions and the experienced interference of others. Simulation results show that the proposed potential game theoretic IA algorithm outperforms the existing algorithm because it minimises the aggregate weighted interference and achieves higher network rate. Copyright (c) 2012 John Wiley & Sons, Ltd.

A Fully Distributed Algorithm for Dynamic Channel Adaptation in Canonical Communication Networks In this letter, a low-complexity, fully distributed algorithm is designed for dynamic channel adaptation in a canonical communication network, where each player can independently update its action without any information exchange. Both the static and dynamic channel environments are studied. The proposed algorithm converges to a set of correlated equilibria with probability one. Moreover, the optimality property of the problem is analyzed. Simulation results demonstrate that the proposed algorithm achieves a near-optimal performance for interference mitigation.

Stochastic Power Adaptation with Multiagent Reinforcement Learning for Cognitive Wireless Mesh Networks As the scarce spectrum resource is becoming overcrowded, cognitive radio indicates great flexibility to improve the spectrum efficiency by opportunistically accessing the authorized frequency bands. One of the critical challenges for operating such radios in a network is how to efficiently allocate transmission powers and frequency resource among the secondary users (SUs) while satisfying the quality-of-service constraints of the primary users. In this paper, we focus on the noncooperative power allocation problem in cognitive wireless mesh networks formed by a number of clusters with the consideration of energy efficiency. Due to the SUs' dynamic and spontaneous properties, the problem is modeled as a stochastic learning process. We first extend the single-agent $(Q)$-learning to a multiuser context, and then propose a conjecture-based multiagent $(Q)$-learning algorithm to achieve the optimal transmission strategies with only private and incomplete information. An intelligent SU performs $(Q)$-function updates based on the conjecture over the other SUs' stochastic behaviors. This learning algorithm provably converges given certain restrictions that arise during the learning procedure. Simulation experiments are used to verify the performance of our algorithm and demonstrate its effectiveness of improving the energy efficiency.

Distributed Resource Allocation for D2D Communications Underlaying Cellular Networks in Time-Varying Environment. In this letter, we address joint channel and power allocation in a device-to-device (D2D) network underlaying a cellular network in a time-varying environment. A fully distributed solution, which does not require information exchange, is proposed to allocate channel and power levels to D2D pairs while ensuring the quality of service (QoS) of the cellular user equipments (CUEs). The problem is modeled as a Stackelberg game with pricing. At the leader level, base station sets prices for the channels to ensure the QoS of the CUEs. At the follower level, D2D pairs use an uncoupled stochastic learning algorithm to learn the channel indices and power levels while minimizing the weighted aggregate interference and the price paid. The follower game is shown to be an ordinal potential game. We perform simulations to study the convergence of the algorithm.

Interference-Aware Online Distributed Channel Selection for Multicluster FANET: A Potential Game Approach The deployment of large-scale UAV clusters for cluster advantages will lead to high competition and excessive congestion of spectrum resources, which results in mutual interference. This paper investigates the interference-aware online spectrum access problem for multicluster Flying Ad-Hoc Network (FANET) under different network topologies, i.e., the locations of UAV clusters are varying. First, the problem is formulated as a data-assisted multistage channel access game with the goal of mitigating interference of all UAV clusters and decreasing channel switching cost during each slot. As we prove that the game is an exact potential game that guarantees at least one pure-strategy Nash equilibrium and interference-aware online channel preserving based concurrent best response (IOCPCBR) algorithm, an online distributed algorithm is proposed to achieve the desirable solution. Finally, the simulation results demonstrate the validity and effectiveness of the multistage channel access game as well as IOCPCBR algorithm.

Maximum ratio transmission This paper presents the concept, principles, and analysis of maximum ratio transmission for wireless communications, where multiple antennas are used for both transmission and reception. The principles and analysis are applicable to general cases, including maximum-ratio combining. Simulation results agree with the analysis. The analysis shows that the average overall signal-to-mise ratio (SNR) is proportional to the cross correlation between channel vectors and that error probability decreases inversely with the (L×K)th power of the average SNR

Long short-term memory. Learning to store information over extended time intervals by recurrent backpropagation takes a very long time, mostly because of insufficient, decaying error backflow. We briefly review Hochreiter's (1991) analysis of this problem, then address it by introducing a novel, efficient, gradient-based method called long short-term memory (LSTM). Truncating the gradient where this does not do harm, LSTM can learn to bridge minimal time lags in excess of 1000 discrete-time steps by enforcing constant error flow through constant error carousels within special units. Multiplicative gate units learn to open and close access to the constant error flow. LSTM is local in space and time; its computational complexity per time step and weight is O(1). Our experiments with artificial data involve local, distributed, real-valued, and noisy pattern representations. In comparisons with real-time recurrent learning, back propagation through time, recurrent cascade correlation, Elman nets, and neural sequence chunking, LSTM leads to many more successful runs, and learns much faster. LSTM also solves complex, artificial long-time-lag tasks that have never been solved by previous recurrent network algorithms.

Joint Task Offloading and Resource Allocation for Multi-Server Mobile-Edge Computing Networks Mobile-edge computing (MEC) is an emerging paradigm that provides a capillary distribution of cloud computing capabilities to the edge of the wireless access network, enabling rich services and applications in close proximity to the end users. In this paper, an MEC enabled multi-cell wireless network is considered where each base station (BS) is equipped with a MEC server that assists mobile users...

Space-time modeling of traffic flow. This paper discusses the application of space-time autoregressive integrated moving average (STARIMA) methodology for representing traffic flow patterns. Traffic flow data are in the form of spatial time series and are collected at specific locations at constant intervals of time. Important spatial characteristics of the space-time process are incorporated in the STARIMA model through the use of weighting matrices estimated on the basis of the distances among the various locations where data are collected. These matrices distinguish the space-time approach from the vector autoregressive moving average (VARMA) methodology and enable the model builders to control the number of the parameters that have to be estimated. The proposed models can be used for short-term forecasting of space-time stationary traffic-flow processes and for assessing the impact of traffic-flow changes on other parts of the network. The three-stage iterative space-time model building procedure is illustrated using 7.5min average traffic flow data for a set of 25 loop-detectors located at roads that direct to the centre of the city of Athens, Greece. Data for two months with different traffic-flow characteristics are modelled in order to determine the stability of the parameter estimation.

A simplified dual neural network for quadratic programming with its KWTA application. The design, analysis, and application of a new recurrent neural network for quadratic programming, called simplified dual neural network, are discussed. The analysis mainly concentrates on the convergence property and the computational complexity of the neural network. The simplified dual neural network is shown to be globally convergent to the exact optimal solution. The complexity of the neural network architecture is reduced with the number of neurons equal to the number of inequality constraints. Its application to k-winners-take-all (KWTA) operation is discussed to demonstrate how to solve problems with this neural network.

G2-type SRMPC scheme for synchronous manipulation of two redundant robot arms. In this paper, to remedy the joint-angle drift phenomenon for manipulation of two redundant robot arms, a novel scheme for simultaneous repetitive motion planning and control (SRMPC) at the joint-acceleration level is proposed, which consists of two subschemes. To do so, the performance index of each SRMPC subscheme is derived and designed by employing the gradient dynamics twice, of which a convergence theorem and its proof are presented. In addition, for improving the accuracy of the motion planning and control, position error, and velocity, error feedbacks are incorporated into the forward kinematics equation and analyzed via Zhang neural-dynamics method. Then the two subschemes are simultaneously reformulated as two quadratic programs (QPs), which are finally unified into one QP problem. Furthermore, a piecewise-linear projection equation-based neural network (PLPENN) is used to solve the unified QP problem, which can handle the strictly convex QP problem in an inverse-free manner. More importantly, via such a unified QP formulation and the corresponding PLPENN solver, the synchronism of two redundant robot arms is guaranteed. Finally, two given tasks are fulfilled by 2 three-link and 2 five-link planar robot arms, respectively. Computer-simulation results validate the efficacy and accuracy of the SRMPC scheme and the corresponding PLPENN solver for synchronous manipulation of two redundant robot arms.

Distributed Kalman consensus filter with event-triggered communication: Formulation and stability analysis. •The problem of distributed state estimation in sensor networks with event-triggered communication schedules on both sensor-to-estimator channel and estimator-to-estimator channel is studied.•An event-triggered KCF is designed by deriving the optimal Kalman gain matrix which minimizes the mean squared error.•A computational scalable form of the proposed filter is presented by some approximations.•An appropriate choice of the consensus gain matrix is provided to ensure the stochastic stability of the proposed filter.

Intention-detection strategies for upper limb exosuits: model-based myoelectric vs dynamic-based control The cognitive human-robot interaction between an exosuit and its wearer plays a key role in determining both the biomechanical effects of the device on movements and its perceived effectiveness. There is a lack of evidence, however, on the comparative performance of different control methods, implemented on the same device. Here, we compare two different control approaches on the same robotic suit: a model-based myoelectric control (myoprocessor), which estimates the joint torque from the activation of target muscles, and a dynamic-based control that provides support against gravity using an inverse dynamic model. Tested on a cohort of four healthy participants, assistance from the exosuit results in a marked reduction in the effort of muscles working against gravity with both control approaches (peak reduction of 68.6±18.8%, for the dynamic arm model and 62.4±25.1% for the myoprocessor), when compared to an unpowered condition. Neither of the two controllers had an affect on the performance of their users in a joint-angle tracking task (peak errors of 15.4° and 16.4° for the dynamic arm model and myoprocessor, respectively, compared to 13.1o in the unpowered condition). However, our results highlight the remarkable adaptability of the myoprocessor to seamlessly adapt to changing external dynamics.

Neural-Dynamic-Method-Based Dual-Arm CMG Scheme With Time-Varying Constraints Applied to Humanoid Robots. We propose a dual-arm cyclic-motion-generation (DACMG) scheme by a neural-dynamic method, which can remedy the joint-angle-drift phenomenon of a humanoid robot. In particular, according to a neural-dynamic design method, first, a cyclic-motion performance index is exploited and applied. This cyclic-motion performance index is then integrated into a quadratic programming (QP)-type scheme with time-...

Taylor O(h³) Discretization of ZNN Models for Dynamic Equality-Constrained Quadratic Programming With Application to Manipulators In this paper, a new Taylor-type numerical differentiation formula is first presented to discretize the continuous-time Zhang neural network (ZNN), and obtain higher computational accuracy. Based on the Taylor-type formula, two Taylor-type discrete-time ZNN models (termed Taylor-type discrete-time ZNNK and Taylor-type discrete-time ZNNU models) are then proposed and discussed to perform online dynamic equality-constrained quadratic programming. For comparison, Euler-type discrete-time ZNN models (called Euler-type discrete-time ZNNK and Euler-type discrete-time ZNNU models) and Newton iteration, with interesting links being found, are also presented. It is proved herein that the steady-state residual errors of the proposed Taylor-type discrete-time ZNN models, Euler-type discrete-time ZNN models, and Newton iteration have the patterns of O(h³), O(h²), and O(h), respectively, with h denoting the sampling gap. Numerical experiments, including the application examples, are carried out, of which the results further substantiate the theoretical findings and the efficacy of Taylor-type discrete-time ZNN models. Finally, the comparisons with Taylor-type discrete-time derivative model and other Lagrange-type discrete-time ZNN models for dynamic equality-constrained quadratic programming substantiate the superiority of the proposed Taylor-type discrete-time ZNN models once again.

Li-function activated ZNN with finite-time convergence applied to redundant-manipulator kinematic control via time-varying Jacobian matrix pseudoinversion. This paper presents and investigates the application of Zhang neural network (ZNN) activated by Li function to kinematic control of redundant robot manipulators via time-varying Jacobian matrix pseudoinversion. That is, by using Li activation function and by computing the time-varying pseudoinverse of the Jacobian matrix (of the robot manipulator), the resultant ZNN model is applied to redundant-manipulator kinematic control. Note that there are nine novelties and differences of ZNN from the conventional gradient neural network in the research methodology. More importantly, such a Li-function activated ZNN (LFAZNN) model has the property of finite-time convergence (showing its feasibility to redundant-manipulator kinematic control). Simulation results based on a four-link planar robot manipulator and a PA10 robot manipulator further demonstrate the effectiveness of the presented LFAZNN model, as well as show the LFAZNN application prospect.

Neural Network and Jacobian Method for Solving the Inverse Statics of a Cable-Driven Soft Arm With Nonconstant Curvature The solution of the inverse kinematics problem of soft manipulators is essential to generate paths in the task space. The inverse kinematics problem of constant curvature or piecewise constant curvature manipulators has already been solved by using different methods, which include closed-form analytical approaches and iterative methods based on the Jacobian method. On the other hand, the inverse kinematics problem of nonconstant curvature manipulators remains unsolved. This study represents one of the first attempts in this direction. It presents both a model-based method and a supervised learning method to solve the inverse statics of nonconstant curvature soft manipulators. In particular, a Jacobian-based method and a feedforward neural network are chosen and tested experimentally. A comparative analysis has been conducted in terms of accuracy and computational time.

Model-Less Hybrid Position/Force Control: A Minimalist Approach for Continuum Manipulators in Unknown, Constrained Environments. Continuum manipulators are designed to operate in constrained environments that are often unknown or unsensed, relying on body compliance to conform to obstacles. The interaction mechanics between the compliant body and unknown environment present significant challenges for traditional robot control techniques based on modeling these interactions exactly. In this letter, we describe a hybrid posit...

Adaptive Fuzzy Backstepping Tracking Control for Flexible Robotic Manipulator In this paper, an adaptive fuzzy state feedback control method is proposed for the single-link robotic manipulator system. The considered system contains unknown nonlinear function and actuator saturation. Fuzzy logic systems (FLSs) and a smooth function are used to approximate the unknown nonlinearities and the actuator saturation, respectively. By combining the command-filter technique with the ...

A simplified dual neural network for quadratic programming with its KWTA application. The design, analysis, and application of a new recurrent neural network for quadratic programming, called simplified dual neural network, are discussed. The analysis mainly concentrates on the convergence property and the computational complexity of the neural network. The simplified dual neural network is shown to be globally convergent to the exact optimal solution. The complexity of the neural network architecture is reduced with the number of neurons equal to the number of inequality constraints. Its application to k-winners-take-all (KWTA) operation is discussed to demonstrate how to solve problems with this neural network.

Event-Triggered Real-Time Scheduling of Stabilizing Control Tasks In this note, we revisit the problem of scheduling stabilizing control tasks on embedded processors. We start from the paradigm that a real-time scheduler could be regarded as a feedback controller that decides which task is executed at any given instant. This controller has for objective guaranteeing that (control unrelated) software tasks meet their deadlines and that stabilizing control tasks asymptotically stabilize the plant. We in- vestigate a simple event-triggered scheduler based on this feedback para- digm and show how it leads to guaranteed performance thus relaxing the more traditional periodic execution requirements. Small embedded microprocessors are quickly becoming an essen- tial part of the most diverse applications. A particularly interesting ex- ample are physically distributed sensor/actuator networks responsible for collecting and processing information, and to react to this infor- mation through actuation. The embedded microprocessors forming the computational core of these networks are required to execute a variety of tasks comprising the relay of information packets in multihop com- munication schemes, monitoring physical quantities, and computation of feedback control laws. Since we are dealing with resource-limited microprocessors, it becomes important to assess to what extent we can increase the functionality of these embedded devices through novel real-time scheduling algorithms based on event-triggered rather than time-triggered execution of control tasks. We investigate in this note a very simple event-triggered scheduling algorithm that preempts running tasks to execute the control task when- ever a certain error becomes large when compared with the state norm. This idea is an adaptation to the scheduling context of several tech- niques used to study problems of control under communication con- straints (6), (11), (18). We take explicitly into account the response time of the control task and show that the proposed scheduling policy guarantees semi-global asymptotical stability. We also show existence of a minimal time between consecutive executions of the control task under the proposed event-triggered scheduling policy. This time, when regarded as a lower bound on release times of the control task, suggests that periodic implementations of control tasks can be relaxed to more flexible aperiodic or event-triggered real-time implementations while guaranteeing desired levels of performance. The proposed approach is illustrated with simulation results. Real-time scheduling of control tasks has received renewed interest from the academic community in the past years (2), (5), (8)-(10), (14), (22). Common to all these approaches, that followed from the initial work of Seto and co-authors (23), is the underlying principle that better control performance is achieved by providing more CPU time to control tasks. This can be accomplished in two different ways: letting control

Neural Architecture Transfer Neural architecture search (NAS) has emerged as a promising avenue for automatically designing task-specific neural networks. Existing NAS approaches require one complete search for each deployment specification of hardware or objective. This is a computationally impractical endeavor given the potentially large number of application scenarios. In this paper, we propose Neural Architecture ...

Metaheuristics in combinatorial optimization: Overview and conceptual comparison The field of metaheuristics for the application to combinatorial optimization problems is a rapidly growing field of research. This is due to the importance of combinatorial optimization problems for the scientific as well as the industrial world. We give a survey of the nowadays most important metaheuristics from a conceptual point of view. We outline the different components and concepts that are used in the different metaheuristics in order to analyze their similarities and differences. Two very important concepts in metaheuristics are intensification and diversification. These are the two forces that largely determine the behavior of a metaheuristic. They are in some way contrary but also complementary to each other. We introduce a framework, that we call the I&D frame, in order to put different intensification and diversification components into relation with each other. Outlining the advantages and disadvantages of different metaheuristic approaches we conclude by pointing out the importance of hybridization of metaheuristics as well as the integration of metaheuristics and other methods for optimization.

Surrogate-assisted evolutionary computation: Recent advances and future challenges Surrogate-assisted, or meta-model based evolutionary computation uses efficient computational models, often known as surrogates or meta-models, for approximating the fitness function in evolutionary algorithms. Research on surrogate-assisted evolutionary computation began over a decade ago and has received considerably increasing interest in recent years. Very interestingly, surrogate-assisted evolutionary computation has found successful applications not only in solving computationally expensive single- or multi-objective optimization problems, but also in addressing dynamic optimization problems, constrained optimization problems and multi-modal optimization problems. This paper provides a concise overview of the history and recent developments in surrogate-assisted evolutionary computation and suggests a few future trends in this research area.

Sliding mode control for uncertain discrete-time systems with Markovian jumping parameters and mixed delays This paper is concerned with the robust sliding mode control (SMC) problem for a class of uncertain discrete-time Markovian jump systems with mixed delays. The mixed delays consist of both the discrete time-varying delays and the infinite distributed delays. The purpose of the addressed problem is to design a sliding mode controller such that, in the simultaneous presence of parameter uncertainties, Markovian jumping parameters and mixed time-delays, the state trajectories are driven onto the pre-defined sliding surface and the resulting sliding mode dynamics is stochastically stable in the mean-square sense. A discrete-time sliding surface is firstly constructed and an SMC law is synthesized to ensure the reaching condition. Moreover, by constructing a new Lyapunov–Krasovskii functional and employing the delay-fractioning approach, a sufficient condition is established to guarantee the stochastic stability of the sliding mode dynamics. Such a condition is characterized in terms of a set of matrix inequalities that can be easily solved by using the semi-definite programming method. A simulation example is given to illustrate the effectiveness and feasibility of the proposed design scheme.

Adversarial Example Generation with Syntactically Controlled Paraphrase Networks. We propose syntactically controlled paraphrase networks (SCPNs) and use them to generate adversarial examples. Given a sentence and a target syntactic form (e.g., a constituency parse), SCPNs are trained to produce a paraphrase of the sentence with the desired syntax. We show it is possible to create training data for this task by first doing backtranslation at a very large scale, and then using a parser to label the syntactic transformations that naturally occur during this process. Such data allows us to train a neural encoder-decoder model with extra inputs to specify the target syntax. A combination of automated and human evaluations show that SCPNs generate paraphrases that follow their target specifications without decreasing paraphrase quality when compared to baseline (uncontrolled) paraphrase systems. Furthermore, they are more capable of generating syntactically adversarial examples that both (1) fool pretrained models and (2) improve the robustness of these models to syntactic variation when used to augment their training data.

A Hierarchical Architecture Using Biased Min-Consensus for USV Path Planning This paper proposes a hierarchical architecture using the biased min-consensus (BMC) method, to solve the path planning problem of unmanned surface vessel (USV). We take the fixed-point monitoring mission as an example, where a series of intermediate monitoring points should be visited once by USV. The whole framework incorporates the low-level layer planning the standard path between any two intermediate points, and the high-level fashion determining their visiting sequence. First, the optimal standard path in terms of voyage time and risk measure is planned by the BMC protocol, given that the corresponding graph is constructed with node state and edge weight. The USV will avoid obstacles or keep a certain distance safely, and arrive at the target point quickly. It is proven theoretically that the state of the graph will converge to be stable after finite iterations, i.e., the optimal solution can be found by BMC with low calculation complexity. Second, by incorporating the constraint of intermediate points, their visiting sequence is optimized by BMC again with the reconstruction of a new virtual graph based on the former planned results. The extensive simulation results in various scenarios also validate the feasibility and effectiveness of our method for autonomous navigation.

Personalized Driver Workload Estimation Using Deep Neural Network Learning From Physiological and Vehicle Signals Drivers often engage in secondary in-vehicle activities that can be functional and/or to relieve monotony. Drivers believe they can safely do so when their perceived workload is low. However, driving requires concurrent execution of cognitive, physical, perceptual and motor tasks. Over allocation of a driver's attention to secondary tasks may impair the driver's control of the vehicle and attention to the surrounding traffic. Accurate assessment of a driver's workload is an important, but challenging, research topic with many applications in intelligent vehicle systems related to driving safety and enhance driving experience. In this article, we present our research on driver workload detection based on driver's physiological, vehicle signals as well as traffic contexts such as congestion level and traffic events. We obtained a collection of data from real driving scenarios. Twenty participants were recruited, whose workload was scaled to “low” as a base level, and “elevated” as a higher level. We developed two convolutional neural networks for multivariate temporal series (MTS-CNN). Extensive experiments were conducted in two scopes: within driver and cross-driver. The within-driver scope concentrates on experiments using data from a single driver, while in the cross-driver scope, transfer learning is leveraged and discussed. The experimental results demonstrate that one of the proposed models, i.e., MTS-CNN2, which combines features captured by the convolutional layers at all levels, is capable of learning well from the combined temporal physiological and vehicle signals and obtains the best performance.

Analysing user physiological responses for affective video summarisation. Video summarisation techniques aim to abstract the most significant content from a video stream. This is typically achieved by processing low-level image, audio and text features which are still quite disparate from the high-level semantics that end users identify with (the ‘semantic gap’). Physiological responses are potentially rich indicators of memorable or emotionally engaging video content for a given user. Consequently, we investigate whether they may serve as a suitable basis for a video summarisation technique by analysing a range of user physiological response measures, specifically electro-dermal response (EDR), respiration amplitude (RA), respiration rate (RR), blood volume pulse (BVP) and heart rate (HR), in response to a range of video content in a variety of genres including horror, comedy, drama, sci-fi and action. We present an analysis framework for processing the user responses to specific sub-segments within a video stream based on percent rank value normalisation. The application of the analysis framework reveals that users respond significantly to the most entertaining video sub-segments in a range of content domains. Specifically, horror content seems to elicit significant EDR, RA, RR and BVP responses, and comedy content elicits comparatively lower levels of EDR, but does seem to elicit significant RA, RR, BVP and HR responses. Drama content seems to elicit less significant physiological responses in general, and both sci-fi and action content seem to elicit significant EDR responses. We discuss the implications this may have for future affective video summarisation approaches.

On the roles of eye gaze and head dynamics in predicting driver's intent to change lanes Driver behavioral cues may present a rich source of information and feedback for future intelligent advanced driver-assistance systems (ADASs). With the design of a simple and robust ADAS in mind, we are interested in determining the most important driver cues for distinguishing driver intent. Eye gaze may provide a more accurate proxy than head movement for determining driver attention, whereas the measurement of head motion is less cumbersome and more reliable in harsh driving conditions. We use a lane-change intent-prediction system (McCall et al., 2007) to determine the relative usefulness of each cue for determining intent. Various combinations of input data are presented to a discriminative classifier, which is trained to output a prediction of probable lane-change maneuver at a particular point in the future. Quantitative results from a naturalistic driving study are presented and show that head motion, when combined with lane position and vehicle dynamics, is a reliable cue for lane-change intent prediction. The addition of eye gaze does not improve performance as much as simpler head dynamics cues. The advantage of head data over eye data is shown to be statistically significant (p

Detection of Driver Fatigue Caused by Sleep Deprivation This paper aims to provide reliable indications of driver drowsiness based on the characteristics of driver-vehicle interaction. A test bed was built under a simulated driving environment, and a total of 12 subjects participated in two experiment sessions requiring different levels of sleep (partial sleep-deprivation versus no sleep-deprivation) before the experiment. The performance of the subjects was analyzed in a series of stimulus-response and routine driving tasks, which revealed the performance differences of drivers under different sleep-deprivation levels. The experiments further demonstrated that sleep deprivation had greater effect on rule-based than on skill-based cognitive functions: when drivers were sleep-deprived, their performance of responding to unexpected disturbances degraded, while they were robust enough to continue the routine driving tasks such as lane tracking, vehicle following, and lane changing. In addition, we presented both qualitative and quantitative guidelines for designing drowsy-driver detection systems in a probabilistic framework based on the paradigm of Bayesian networks. Temporal aspects of drowsiness and individual differences of subjects were addressed in the framework.

Online Prediction of Driver Distraction Based on Brain Activity Patterns This paper presents a new computational framework for early detection of driver distractions (map viewing) using brain activity measured by electroencephalographic (EEG) signals. Compared with most studies in the literature, which are mainly focused on the classification of distracted and nondistracted periods, this study proposes a new framework to prospectively predict the start and end of a distraction period, defined by map viewing. The proposed prediction algorithm was tested on a data set of continuous EEG signals recorded from 24 subjects. During the EEG recordings, the subjects were asked to drive from an initial position to a destination using a city map in a simulated driving environment. The overall accuracy values for the prediction of the start and the end of map viewing were 81% and 70%, respectively. The experimental results demonstrated that the proposed algorithm can predict the start and end of map viewing with relatively high accuracy and can be generalized to individual subjects. The outcome of this study has a high potential to improve the design of future intelligent navigation systems. Prediction of the start of map viewing can be used to provide route information based on a driver's needs and consequently avoid map-viewing activities. Prediction of the end of map viewing can be used to provide warnings for potential long map-viewing durations. Further development of the proposed framework and its applications in driver-distraction predictions are also discussed.

Keep Your Scanners Peeled: Gaze Behavior as a Measure of Automation Trust During Highly Automated Driving. Objective: The feasibility of measuring drivers' automation trust via gaze behavior during highly automated driving was assessed with eye tracking and validated with self-reported automation trust in a driving simulator study. Background: Earlier research from other domains indicates that drivers' automation trust might be inferred from gaze behavior, such as monitoring frequency. Method: The gaze behavior and self-reported automation trust of 35 participants attending to a visually demanding non-driving-related task (NDRT) during highly automated driving was evaluated. The relationship between dispositional, situational, and learned automation trust with gaze behavior was compared. Results: Overall, there was a consistent relationship between drivers' automation trust and gaze behavior. Participants reporting higher automation trust tended to monitor the automation less frequently. Further analyses revealed that higher automation trust was associated with lower monitoring frequency of the automation during NDRTs, and an increase in trust over the experimental session was connected with a decrease in monitoring frequency. Conclusion: We suggest that (a) the current results indicate a negative relationship between drivers' self-reported automation trust and monitoring frequency, (b) gaze behavior provides a more direct measure of automation trust than other behavioral measures, and (c) with further refinement, drivers' automation trust during highly automated driving might be inferred from gaze behavior. Application: Potential applications of this research include the estimation of drivers' automation trust and reliance during highly automated driving.

3D separable convolutional neural network for dynamic hand gesture recognition. •The Frame Difference method is used to pre-process the input in order to filter the background.•A 3D separable CNN is proposed for dynamic gesture recognition. The standard 3D convolution process is decomposed into two processes: 3D depth-wise and 3D point-wise.•By the application of skip connection and layer-wise learning rate, the undesirable gradient dispersion due to the separation operation is solved and the performance of the network is improved.•A dynamic hand gesture library is built through HoloLens.

Deep convolutional neural network-based Bernoulli heatmap for head pose estimation Head pose estimation is a crucial problem for many tasks, such as driver attention, fatigue detection, and human behaviour analysis. It is well known that neural networks are better at handling classification problems than regression problems. It is an extremely nonlinear process to let the network output the angle value directly for optimization learning, and the weight constraint of the loss function will be relatively weak. This paper proposes a novel Bernoulli heatmap for head pose estimation from a single RGB image. Our method can achieve the positioning of the head area while estimating the angles of the head. The Bernoulli heatmap makes it possible to construct fully convolutional neural networks without fully connected layers and provides a new idea for the output form of head pose estimation. A deep convolutional neural network (CNN) structure with multiscale representations is adopted to maintain high-resolution information and low-resolution information in parallel. This kind of structure can maintain rich, high-resolution representations. In addition, channelwise fusion is adopted to make the fusion weights learnable instead of simple addition with equal weights. As a result, the estimation is spatially more precise and potentially more accurate. The effectiveness of the proposed method is empirically demonstrated by comparing it with other state-of-the-art methods on public datasets.

Reinforcement learning based data fusion method for multi-sensors In order to improve detection system robustness and reliability, multi-sensors fusion is used in modern air combat. In this paper, a data fusion method based on reinforcement learning is developed for multi-sensors. Initially, the cubic B-spline interpolation is used to solve time alignment problems of multisource data. Then, the reinforcement learning based data fusion (RLBDF) method is proposed to obtain the fusion results. With the case that the priori knowledge of target is obtained, the fusion accuracy reinforcement is realized by the error between fused value and actual value. Furthermore, the Fisher information is instead used as the reward if the priori knowledge is unable to be obtained. Simulations results verify that the developed method is feasible and effective for the multi-sensors data fusion in air combat.

Mobile Edge Computing Enabled 5G Health Monitoring for Internet of Medical Things: A Decentralized Game Theoretic Approach The prompt evolution of Internet of Medical Things (IoMT) promotes pervasive in-home health monitoring networks. However, excessive requirements of patients result in insufficient spectrum resources and communication overload. Mobile Edge Computing (MEC) enabled 5G health monitoring is conceived as a favorable paradigm to tackle such an obstacle. In this paper, we construct a cost-efficient in-home health monitoring system for IoMT by dividing it into two sub-networks, i.e., intra-Wireless Body Area Networks (WBANs) and beyond-WBANs. Highlighting the characteristics of IoMT, the cost of patients depends on medical criticality, Age of Information (AoI) and energy consumption. For intra-WBANs, a cooperative game is formulated to allocate the wireless channel resources. While for beyond-WBANs, considering the individual rationality and potential selfishness, a decentralized non-cooperative game is proposed to minimize the system-wide cost in IoMT. We prove that the proposed algorithm can reach a Nash equilibrium. In addition, the upper bound of the algorithm time complexity and the number of patients benefiting from MEC is theoretically derived. Performance evaluations demonstrate the effectiveness of our proposed algorithm with respect to the system-wide cost and the number of patients benefiting from MEC.

Artificial fish swarm algorithm: a survey of the state-of-the-art, hybridization, combinatorial and indicative applications FSA (artificial fish-swarm algorithm) is one of the best methods of optimization among the swarm intelligence algorithms. This algorithm is inspired by the collective movement of the fish and their various social behaviors. Based on a series of instinctive behaviors, the fish always try to maintain their colonies and accordingly demonstrate intelligent behaviors. Searching for food, immigration and dealing with dangers all happen in a social form and interactions between all fish in a group will result in an intelligent social behavior.This algorithm has many advantages including high convergence speed, flexibility, fault tolerance and high accuracy. This paper is a review of AFSA algorithm and describes the evolution of this algorithm along with all improvements, its combination with various methods as well as its applications. There are many optimization methods which have a affinity with this method and the result of this combination will improve the performance of this method. Its disadvantages include high time complexity, lack of balance between global and local search, in addition to lack of benefiting from the experiences of group members for the next movements.

Short-Term Traffic Flow Forecasting: An Experimental Comparison of Time-Series Analysis and Supervised Learning The literature on short-term traffic flow forecasting has undergone great development recently. Many works, describing a wide variety of different approaches, which very often share similar features and ideas, have been published. However, publications presenting new prediction algorithms usually employ different settings, data sets, and performance measurements, making it difficult to infer a clear picture of the advantages and limitations of each model. The aim of this paper is twofold. First, we review existing approaches to short-term traffic flow forecasting methods under the common view of probabilistic graphical models, presenting an extensive experimental comparison, which proposes a common baseline for their performance analysis and provides the infrastructure to operate on a publicly available data set. Second, we present two new support vector regression models, which are specifically devised to benefit from typical traffic flow seasonality and are shown to represent an interesting compromise between prediction accuracy and computational efficiency. The SARIMA model coupled with a Kalman filter is the most accurate model; however, the proposed seasonal support vector regressor turns out to be highly competitive when performing forecasts during the most congested periods.

TSCA: A Temporal-Spatial Real-Time Charging Scheduling Algorithm for On-Demand Architecture in Wireless Rechargeable Sensor Networks. The collaborative charging issue in Wireless Rechargeable Sensor Networks (WRSNs) is a popular research problem. With the help of wireless power transfer technology, electrical energy can be transferred from wireless charging vehicles (WCVs) to sensors, providing a new paradigm to prolong network lifetime. Existing techniques on collaborative charging usually take the periodical and deterministic approach, but neglect influences of non-deterministic factors such as topological changes and node failures, making them unsuitable for large-scale WRSNs. In this paper, we develop a temporal-spatial charging scheduling algorithm, namely TSCA, for the on-demand charging architecture. We aim to minimize the number of dead nodes while maximizing energy efficiency to prolong network lifetime. First, after gathering charging requests, a WCV will compute a feasible movement solution. A basic path planning algorithm is then introduced to adjust the charging order for better efficiency. Furthermore, optimizations are made in a global level. Then, a node deletion algorithm is developed to remove low efficient charging nodes. Lastly, a node insertion algorithm is executed to avoid the death of abandoned nodes. Extensive simulations show that, compared with state-of-the-art charging scheduling algorithms, our scheme can achieve promising performance in charging throughput, charging efficiency, and other performance metrics.

A novel adaptive dynamic programming based on tracking error for nonlinear discrete-time systems In this paper, to eliminate the tracking error by using adaptive dynamic programming (ADP) algorithms, a novel formulation of the value function is presented for the optimal tracking problem (TP) of nonlinear discrete-time systems. Unlike existing ADP methods, this formulation introduces the control input into the tracking error, and ignores the quadratic form of the control input directly, which makes the boundedness and convergence of the value function independent of the discount factor. Based on the proposed value function, the optimal control policy can be deduced without considering the reference control input. Value iteration (VI) and policy iteration (PI) methods are applied to prove the optimality of the obtained control policy, and derived the monotonicity property and convergence of the iterative value function. Simulation examples realized with neural networks and the actor–critic structure are provided to verify the effectiveness of the proposed ADP algorithm.

Hybrid algorithm optimization for coverage problem in wireless sensor networks With the continuous development of evolutionary computing, many excellent algorithms have emerged, which are applied in all walks of life to solve various practical problems. In this paper, two hybrid fish, bird and insect algorithms based on different architectures are proposed to solve the optimal coverage problem in wireless sensor networks. The algorithm combines the characteristics of three algorithms, namely, particle swarm optimization algorithm, Phasmatodea population evolution algorithm and fish migration optimization algorithm. The new algorithm has the advantages of the three algorithms. In order to prove the effectiveness of the algorithm, we first test it on 28 benchmark functions. The results show that the two hybrid fish, bird and insect algorithms with different architectures have significant advantages. Then we apply the proposed algorithm to solve the coverage problem of wireless sensor networks through experimental simulation. The experimental results show the advantages of our proposed algorithm and prove that our proposed hybrid fish, bird and insect algorithm is suitable for solving the coverage problem of wireless sensor networks.

The Sybil Attack Large-scale peer-to-peer systems facesecurity threats from faulty or hostile remotecomputing elements. To resist these threats, manysuch systems employ redundancy. However, if asingle faulty entity can present multiple identities,it can control a substantial fraction of the system,thereby undermining this redundancy. Oneapproach to preventing these &quot;Sybil attacks&quot; is tohave a trusted agency certify identities. Thispaper shows that, without a logically centralizedauthority, Sybil...

BLEU: a method for automatic evaluation of machine translation Human evaluations of machine translation are extensive but expensive. Human evaluations can take months to finish and involve human labor that can not be reused. We propose a method of automatic machine translation evaluation that is quick, inexpensive, and language-independent, that correlates highly with human evaluation, and that has little marginal cost per run. We present this method as an automated understudy to skilled human judges which substitutes for them when there is need for quick or frequent evaluations.

Computational thinking Summary form only given. My vision for the 21st century, Computational Thinking, will be a fundamental skill used by everyone in the world. To reading, writing, and arithmetic, we should add computational thinking to every child's analytical ability. Computational thinking involves solving problems, designing systems, and understanding human behavior by drawing on the concepts fundamental to computer science. Thinking like a computer scientist means more than being able to program a computer. It requires the ability to abstract and thus to think at multiple levels of abstraction. In this talk I will give many examples of computational thinking, argue that it has already influenced other disciplines, and promote the idea that teaching computational thinking can not only inspire future generations to enter the field of computer science but benefit people in all fields.

Fuzzy logic in control systems: fuzzy logic controller. I.

Switching between stabilizing controllers This paper deals with the problem of switching between several linear time-invariant (LTI) controllers—all of them capable of stabilizing a speci4c LTI process—in such a way that the stability of the closed-loop system is guaranteed for any switching sequence. We show that it is possible to 4nd realizations for any given family of controller transfer matrices so that the closed-loop system remains stable, no matter how we switch among the controller. The motivation for this problem is the control of complex systems where con8icting requirements make a single LTI controller unsuitable. ? 2002 Published by Elsevier Science Ltd.

Tabu Search - Part I

Bidirectional recurrent neural networks In the first part of this paper, a regular recurrent neural network (RNN) is extended to a bidirectional recurrent neural network (BRNN). The BRNN can be trained without the limitation of using input information just up to a preset future frame. This is accomplished by training it simultaneously in positive and negative time direction. Structure and training procedure of the proposed network are explained. In regression and classification experiments on artificial data, the proposed structure gives better results than other approaches. For real data, classification experiments for phonemes from the TIMIT database show the same tendency. In the second part of this paper, it is shown how the proposed bidirectional structure can be easily modified to allow efficient estimation of the conditional posterior probability of complete symbol sequences without making any explicit assumption about the shape of the distribution. For this part, experiments on real data are reported

An intensive survey of fair non-repudiation protocols With the phenomenal growth of the Internet and open networks in general, security services, such as non-repudiation, become crucial to many applications. Non-repudiation services must ensure that when Alice sends some information to Bob over a network, neither Alice nor Bob can deny having participated in a part or the whole of this communication. Therefore a fair non-repudiation protocol has to generate non-repudiation of origin evidences intended to Bob, and non-repudiation of receipt evidences destined to Alice. In this paper, we clearly define the properties a fair non-repudiation protocol must respect, and give a survey of the most important non-repudiation protocols without and with trusted third party (TTP). For the later ones we discuss the evolution of the TTP's involvement and, between others, describe the most recent protocol using a transparent TTP. We also discuss some ad-hoc problems related to the management of non-repudiation evidences.

Dynamic movement and positioning of embodied agents in multiparty conversations For embodied agents to engage in realistic multiparty conversation, they must stand in appropriate places with respect to other agents and the environment. When these factors change, such as an agent joining the conversation, the agents must dynamically move to a new location and/or orientation to accommodate. This paper presents an algorithm for simulating movement of agents based on observed human behavior using techniques developed for pedestrian movement in crowd simulations. We extend a previous group conversation simulation to include an agent motion algorithm. We examine several test cases and show how the simulation generates results that mirror real-life conversation settings.

An improved genetic algorithm with conditional genetic operators and its application to set-covering problem The genetic algorithm (GA) is a popular, biologically inspired optimization method. However, in the GA there is no rule of thumb to design the GA operators and select GA parameters. Instead, trial-and-error has to be applied. In this paper we present an improved genetic algorithm in which crossover and mutation are performed conditionally instead of probability. Because there are no crossover rate and mutation rate to be selected, the proposed improved GA can be more easily applied to a problem than the conventional genetic algorithms. The proposed improved genetic algorithm is applied to solve the set-covering problem. Experimental studies show that the improved GA produces better results over the conventional one and other methods.

Lane-level traffic estimations using microscopic traffic variables This paper proposes a novel inference method to estimate lane-level traffic flow, time occupancy and vehicle inter-arrival time on road segments where local information could not be measured and assessed directly. The main contributions of the proposed method are 1) the ability to perform lane-level estimations of traffic flow, time occupancy and vehicle inter-arrival time and 2) the ability to adapt to different traffic regimes by assessing only microscopic traffic variables. We propose a modified Kriging estimation model which explicitly takes into account both spatial and temporal variability. Performance evaluations are conducted using real-world data under different traffic regimes and it is shown that the proposed method outperforms a Kalman filter-based approach.

Convolutional Neural Network-Based Classification of Driver's Emotion during Aggressive and Smooth Driving Using Multi-Modal Camera Sensors. Because aggressive driving often causes large-scale loss of life and property, techniques for advance detection of adverse driver emotional states have become important for the prevention of aggressive driving behaviors. Previous studies have primarily focused on systems for detecting aggressive driver emotion via smart-phone accelerometers and gyro-sensors, or they focused on methods of detecting physiological signals using electroencephalography (EEG) or electrocardiogram (ECG) sensors. Because EEG and ECG sensors cause discomfort to drivers and can be detached from the driver's body, it becomes difficult to focus on bio-signals to determine their emotional state. Gyro-sensors and accelerometers depend on the performance of GPS receivers and cannot be used in areas where GPS signals are blocked. Moreover, if driving on a mountain road with many quick turns, a driver's emotional state can easily be misrecognized as that of an aggressive driver. To resolve these problems, we propose a convolutional neural network (CNN)-based method of detecting emotion to identify aggressive driving using input images of the driver's face, obtained using near-infrared (NIR) light and thermal camera sensors. In this research, we conducted an experiment using our own database, which provides a high classification accuracy for detecting driver emotion leading to either aggressive or smooth (i.e., relaxed) driving. Our proposed method demonstrates better performance than existing methods.

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

Hybrid-Learning-Based Driver Steering Intention Prediction Using Neuromuscular Dynamics The emerging automated driving technology poses a new challenge to driver-automation collaboration, which requires a mutual understanding between humans and machines through their intention identifications. In this article, oriented by human–machine mutual understanding, a driver steering intention prediction method is proposed to better understand human driver&#39;s expectation during driver–vehicle ...

Analysing user physiological responses for affective video summarisation. Video summarisation techniques aim to abstract the most significant content from a video stream. This is typically achieved by processing low-level image, audio and text features which are still quite disparate from the high-level semantics that end users identify with (the ‘semantic gap’). Physiological responses are potentially rich indicators of memorable or emotionally engaging video content for a given user. Consequently, we investigate whether they may serve as a suitable basis for a video summarisation technique by analysing a range of user physiological response measures, specifically electro-dermal response (EDR), respiration amplitude (RA), respiration rate (RR), blood volume pulse (BVP) and heart rate (HR), in response to a range of video content in a variety of genres including horror, comedy, drama, sci-fi and action. We present an analysis framework for processing the user responses to specific sub-segments within a video stream based on percent rank value normalisation. The application of the analysis framework reveals that users respond significantly to the most entertaining video sub-segments in a range of content domains. Specifically, horror content seems to elicit significant EDR, RA, RR and BVP responses, and comedy content elicits comparatively lower levels of EDR, but does seem to elicit significant RA, RR, BVP and HR responses. Drama content seems to elicit less significant physiological responses in general, and both sci-fi and action content seem to elicit significant EDR responses. We discuss the implications this may have for future affective video summarisation approaches.

On the roles of eye gaze and head dynamics in predicting driver's intent to change lanes Driver behavioral cues may present a rich source of information and feedback for future intelligent advanced driver-assistance systems (ADASs). With the design of a simple and robust ADAS in mind, we are interested in determining the most important driver cues for distinguishing driver intent. Eye gaze may provide a more accurate proxy than head movement for determining driver attention, whereas the measurement of head motion is less cumbersome and more reliable in harsh driving conditions. We use a lane-change intent-prediction system (McCall et al., 2007) to determine the relative usefulness of each cue for determining intent. Various combinations of input data are presented to a discriminative classifier, which is trained to output a prediction of probable lane-change maneuver at a particular point in the future. Quantitative results from a naturalistic driving study are presented and show that head motion, when combined with lane position and vehicle dynamics, is a reliable cue for lane-change intent prediction. The addition of eye gaze does not improve performance as much as simpler head dynamics cues. The advantage of head data over eye data is shown to be statistically significant (p

Detection of Driver Fatigue Caused by Sleep Deprivation This paper aims to provide reliable indications of driver drowsiness based on the characteristics of driver-vehicle interaction. A test bed was built under a simulated driving environment, and a total of 12 subjects participated in two experiment sessions requiring different levels of sleep (partial sleep-deprivation versus no sleep-deprivation) before the experiment. The performance of the subjects was analyzed in a series of stimulus-response and routine driving tasks, which revealed the performance differences of drivers under different sleep-deprivation levels. The experiments further demonstrated that sleep deprivation had greater effect on rule-based than on skill-based cognitive functions: when drivers were sleep-deprived, their performance of responding to unexpected disturbances degraded, while they were robust enough to continue the routine driving tasks such as lane tracking, vehicle following, and lane changing. In addition, we presented both qualitative and quantitative guidelines for designing drowsy-driver detection systems in a probabilistic framework based on the paradigm of Bayesian networks. Temporal aspects of drowsiness and individual differences of subjects were addressed in the framework.

Online Prediction of Driver Distraction Based on Brain Activity Patterns This paper presents a new computational framework for early detection of driver distractions (map viewing) using brain activity measured by electroencephalographic (EEG) signals. Compared with most studies in the literature, which are mainly focused on the classification of distracted and nondistracted periods, this study proposes a new framework to prospectively predict the start and end of a distraction period, defined by map viewing. The proposed prediction algorithm was tested on a data set of continuous EEG signals recorded from 24 subjects. During the EEG recordings, the subjects were asked to drive from an initial position to a destination using a city map in a simulated driving environment. The overall accuracy values for the prediction of the start and the end of map viewing were 81% and 70%, respectively. The experimental results demonstrated that the proposed algorithm can predict the start and end of map viewing with relatively high accuracy and can be generalized to individual subjects. The outcome of this study has a high potential to improve the design of future intelligent navigation systems. Prediction of the start of map viewing can be used to provide route information based on a driver's needs and consequently avoid map-viewing activities. Prediction of the end of map viewing can be used to provide warnings for potential long map-viewing durations. Further development of the proposed framework and its applications in driver-distraction predictions are also discussed.

Keep Your Scanners Peeled: Gaze Behavior as a Measure of Automation Trust During Highly Automated Driving. Objective: The feasibility of measuring drivers' automation trust via gaze behavior during highly automated driving was assessed with eye tracking and validated with self-reported automation trust in a driving simulator study. Background: Earlier research from other domains indicates that drivers' automation trust might be inferred from gaze behavior, such as monitoring frequency. Method: The gaze behavior and self-reported automation trust of 35 participants attending to a visually demanding non-driving-related task (NDRT) during highly automated driving was evaluated. The relationship between dispositional, situational, and learned automation trust with gaze behavior was compared. Results: Overall, there was a consistent relationship between drivers' automation trust and gaze behavior. Participants reporting higher automation trust tended to monitor the automation less frequently. Further analyses revealed that higher automation trust was associated with lower monitoring frequency of the automation during NDRTs, and an increase in trust over the experimental session was connected with a decrease in monitoring frequency. Conclusion: We suggest that (a) the current results indicate a negative relationship between drivers' self-reported automation trust and monitoring frequency, (b) gaze behavior provides a more direct measure of automation trust than other behavioral measures, and (c) with further refinement, drivers' automation trust during highly automated driving might be inferred from gaze behavior. Application: Potential applications of this research include the estimation of drivers' automation trust and reliance during highly automated driving.

3D separable convolutional neural network for dynamic hand gesture recognition. •The Frame Difference method is used to pre-process the input in order to filter the background.•A 3D separable CNN is proposed for dynamic gesture recognition. The standard 3D convolution process is decomposed into two processes: 3D depth-wise and 3D point-wise.•By the application of skip connection and layer-wise learning rate, the undesirable gradient dispersion due to the separation operation is solved and the performance of the network is improved.•A dynamic hand gesture library is built through HoloLens.

Cooperative channel assignment for VANETs based on multiagent reinforcement learning. Dynamic channel assignment (DCA) plays a key role in extending vehicular ad-hoc network capacity and mitigating congestion. However, channel assignment under vehicular direct communication scenarios faces mutual influence of large-scale nodes, the lack of centralized coordination, unknown global state information, and other challenges. To solve this problem, a multiagent reinforcement learning (RL) based cooperative DCA (RL-CDCA) mechanism is proposed. Specifically, each vehicular node can successfully learn the proper strategies of channel selection and backoff adaptation from the real-time channel state information (CSI) using two cooperative RL models. In addition, neural networks are constructed as nonlinear Q-function approximators, which facilitates the mapping of the continuously sensed input to the mixed policy output. Nodes are driven to locally share and incorporate their individual rewards such that they can optimize their policies in a distributed collaborative manner. Simulation results show that the proposed multiagent RL-CDCA can better reduce the one-hop packet delay by no less than 73.73%, improve the packet delivery ratio by no less than 12.66% on average in a highly dense situation, and improve the fairness of the global network resource allocation.

Reinforcement learning based data fusion method for multi-sensors In order to improve detection system robustness and reliability, multi-sensors fusion is used in modern air combat. In this paper, a data fusion method based on reinforcement learning is developed for multi-sensors. Initially, the cubic B-spline interpolation is used to solve time alignment problems of multisource data. Then, the reinforcement learning based data fusion (RLBDF) method is proposed to obtain the fusion results. With the case that the priori knowledge of target is obtained, the fusion accuracy reinforcement is realized by the error between fused value and actual value. Furthermore, the Fisher information is instead used as the reward if the priori knowledge is unable to be obtained. Simulations results verify that the developed method is feasible and effective for the multi-sensors data fusion in air combat.

Mobile Edge Computing Enabled 5G Health Monitoring for Internet of Medical Things: A Decentralized Game Theoretic Approach The prompt evolution of Internet of Medical Things (IoMT) promotes pervasive in-home health monitoring networks. However, excessive requirements of patients result in insufficient spectrum resources and communication overload. Mobile Edge Computing (MEC) enabled 5G health monitoring is conceived as a favorable paradigm to tackle such an obstacle. In this paper, we construct a cost-efficient in-home health monitoring system for IoMT by dividing it into two sub-networks, i.e., intra-Wireless Body Area Networks (WBANs) and beyond-WBANs. Highlighting the characteristics of IoMT, the cost of patients depends on medical criticality, Age of Information (AoI) and energy consumption. For intra-WBANs, a cooperative game is formulated to allocate the wireless channel resources. While for beyond-WBANs, considering the individual rationality and potential selfishness, a decentralized non-cooperative game is proposed to minimize the system-wide cost in IoMT. We prove that the proposed algorithm can reach a Nash equilibrium. In addition, the upper bound of the algorithm time complexity and the number of patients benefiting from MEC is theoretically derived. Performance evaluations demonstrate the effectiveness of our proposed algorithm with respect to the system-wide cost and the number of patients benefiting from MEC.

Artificial fish swarm algorithm: a survey of the state-of-the-art, hybridization, combinatorial and indicative applications FSA (artificial fish-swarm algorithm) is one of the best methods of optimization among the swarm intelligence algorithms. This algorithm is inspired by the collective movement of the fish and their various social behaviors. Based on a series of instinctive behaviors, the fish always try to maintain their colonies and accordingly demonstrate intelligent behaviors. Searching for food, immigration and dealing with dangers all happen in a social form and interactions between all fish in a group will result in an intelligent social behavior.This algorithm has many advantages including high convergence speed, flexibility, fault tolerance and high accuracy. This paper is a review of AFSA algorithm and describes the evolution of this algorithm along with all improvements, its combination with various methods as well as its applications. There are many optimization methods which have a affinity with this method and the result of this combination will improve the performance of this method. Its disadvantages include high time complexity, lack of balance between global and local search, in addition to lack of benefiting from the experiences of group members for the next movements.

Short-Term Traffic Flow Forecasting: An Experimental Comparison of Time-Series Analysis and Supervised Learning The literature on short-term traffic flow forecasting has undergone great development recently. Many works, describing a wide variety of different approaches, which very often share similar features and ideas, have been published. However, publications presenting new prediction algorithms usually employ different settings, data sets, and performance measurements, making it difficult to infer a clear picture of the advantages and limitations of each model. The aim of this paper is twofold. First, we review existing approaches to short-term traffic flow forecasting methods under the common view of probabilistic graphical models, presenting an extensive experimental comparison, which proposes a common baseline for their performance analysis and provides the infrastructure to operate on a publicly available data set. Second, we present two new support vector regression models, which are specifically devised to benefit from typical traffic flow seasonality and are shown to represent an interesting compromise between prediction accuracy and computational efficiency. The SARIMA model coupled with a Kalman filter is the most accurate model; however, the proposed seasonal support vector regressor turns out to be highly competitive when performing forecasts during the most congested periods.

TSCA: A Temporal-Spatial Real-Time Charging Scheduling Algorithm for On-Demand Architecture in Wireless Rechargeable Sensor Networks. The collaborative charging issue in Wireless Rechargeable Sensor Networks (WRSNs) is a popular research problem. With the help of wireless power transfer technology, electrical energy can be transferred from wireless charging vehicles (WCVs) to sensors, providing a new paradigm to prolong network lifetime. Existing techniques on collaborative charging usually take the periodical and deterministic approach, but neglect influences of non-deterministic factors such as topological changes and node failures, making them unsuitable for large-scale WRSNs. In this paper, we develop a temporal-spatial charging scheduling algorithm, namely TSCA, for the on-demand charging architecture. We aim to minimize the number of dead nodes while maximizing energy efficiency to prolong network lifetime. First, after gathering charging requests, a WCV will compute a feasible movement solution. A basic path planning algorithm is then introduced to adjust the charging order for better efficiency. Furthermore, optimizations are made in a global level. Then, a node deletion algorithm is developed to remove low efficient charging nodes. Lastly, a node insertion algorithm is executed to avoid the death of abandoned nodes. Extensive simulations show that, compared with state-of-the-art charging scheduling algorithms, our scheme can achieve promising performance in charging throughput, charging efficiency, and other performance metrics.

A novel adaptive dynamic programming based on tracking error for nonlinear discrete-time systems In this paper, to eliminate the tracking error by using adaptive dynamic programming (ADP) algorithms, a novel formulation of the value function is presented for the optimal tracking problem (TP) of nonlinear discrete-time systems. Unlike existing ADP methods, this formulation introduces the control input into the tracking error, and ignores the quadratic form of the control input directly, which makes the boundedness and convergence of the value function independent of the discount factor. Based on the proposed value function, the optimal control policy can be deduced without considering the reference control input. Value iteration (VI) and policy iteration (PI) methods are applied to prove the optimality of the obtained control policy, and derived the monotonicity property and convergence of the iterative value function. Simulation examples realized with neural networks and the actor–critic structure are provided to verify the effectiveness of the proposed ADP algorithm.

A recurrent attention and interaction model for pedestrian trajectory prediction The movement of pedestrians involves temporal continuity, spatial interactivity, and random diversity. As a result, pedestrian trajectory prediction is rather challenging. Most existing trajectory prediction methods tend to focus on just one aspect of these challenges, ignoring the temporal information of the trajectory and making too many assumptions. In this paper, we propose a recurrent attention and interaction (RAI) model to predict pedestrian trajectories. The RAI model consists of a temporal attention module, spatial pooling module, and randomness modeling module. The temporal attention module is proposed to assign different weights to the input sequence of a target, and reduce the speed deviation of different pedestrians. The spatial pooling module is proposed to model not only the social information of neighbors in historical frames, but also the intention of neighbors in the current time. The randomness modeling module is proposed to model the uncertainty and diversity of trajectories by introducing random noise. We conduct extensive experiments on several public datasets. The results demonstrate that our method outperforms many that are state-of-the-art.

Parallel Ecology for Intelligent and Smart Cyber–Physical–Social Systems Welcome to the last issue of IEEE Transactions on Computational Social Systems (TCSS) this year. This is also the last time I am serving as the Editor-in-Chief of this great journal, and I would like to take this opportunity to thank you all for your great help and support during the last three and half years. A special thank must go to my “Executive Editorial Task Force”: Professor Yong Yuan of People's University of China; Drs. Rui Qin, Xiao Wang, and Xueliang Zhao of The State Key Laboratory for Management and Control of Complex Systems, Institute of Automation, Chinese Academy of Sciences; and Dr. Juanjuan Li of Beijing Institute of Technology, for their hardworking and dedication within my term as EiC. We have rejuvenated TCSS in a very short period and maintained its state of healthy growth, and this is a great team I must remember and will be proud of.

Real-time Localization in Outdoor Environments using Stereo Vision and Inexpensive GPS We describe a real-time, low-cost system to localize a mobile robot in outdoor environments. Our system relies on stereo vision to robustly estimate frame-to-frame motion in real time (also known as visual odometry). The motion estimation problem is formulated efficiently in the disparity space and results in accurate and robust estimates of the motion even for a small-baseline configuration. Our system uses inertial measurements to fill in motion estimates when visual odometry fails. This incremental motion is then fused with a low-cost GPS sensor using a Kalman Filter to prevent long-term drifts. Experimental results are presented for outdoor localization in moderately sized environments (\geqslant 100 meters)

Environment-aware Development of Robust Vision-based Cooperative Perception Systems Autonomous vehicles need a complete and robust perception of their environment to correctly understand the surrounding traffic scene and come to the right decisions. Making use of vehicle-to-vehicle (V2V) communication can improve the perception capabilities of autonomous vehicles by extending the range of their own local sensors. For the development of robust cooperative perception systems it is necessary to include varying environmental conditions to the scenarios used for validation. In this paper we present a new approach to investigate a cooperative perception pipeline within simulation under varying rain conditions. We demonstrate our approach on the example of a complete vision-based cooperative perception pipeline. Scenarios with a varying number of cooperative vehicles under different synthetically generated rain variations are used to show the influence of rain on local and cooperative perception.

A Multi-Stream Feature Fusion Approach for Traffic Prediction Accurate and timely traffic flow prediction is crucial for intelligent transportation systems (ITS). Recent advances in graph-based neural networks have achieved promising prediction results. However, some challenges remain, especially regarding graph construction and the time complexity of models. In this paper, we propose a multi-stream feature fusion approach to extract and integrate rich features from traffic data and leverage a data-driven adjacent matrix instead of the distance-based matrix to construct graphs. We calculate the Spearman rank correlation coefficient between monitor stations to obtain the initial adjacent matrix and fine-tune it while training. As to the model, we construct a multi-stream feature fusion block (MFFB) module, which includes a three-channel network and the soft-attention mechanism. The three-channel networks are graph convolutional neural network (GCN), gated recurrent unit (GRU) and fully connected neural network (FNN), which are used to extract spatial, temporal and other features, respectively. The soft-attention mechanism is utilized to integrate the obtained features. The MFFB modules are stacked, and a fully connected layer and a convolutional layer are used to make predictions. We conduct experiments on two real-world traffic prediction tasks and verify that our proposed approach outperforms the state-of-the-art methods within an acceptable time complexity.

GNSS NLOS Exclusion Based on Dynamic Object Detection Using LiDAR Point Cloud Absolute positioning is an essential factor for the arrival of autonomous driving. At present, GNSS is the indispensable source that can supply initial positioning in the commonly used high definition map-based LiDAR point cloud positioning solution for autonomous driving. However, the non-light-of-sight (NLOS) reception dominates GNSS positioning performance in super-urbanized areas. The recent proposed 3D map aided (3DMA) GNSS can mitigate the majority of the NLOS caused by buildings. However, the same phenomenon caused by moving objects in urban areas is currently not modeled in the 3D geographic information system (GIS). Therefore, we present a novel method to exclude the NLOS receptions caused by a double-decker bus, one of the symbolic tall moving objects in road transportations. To estimate the dimension and orientation of the double-decker buses relative to the GNSS receiver, LiDAR-based perception is utilized. By projecting the relative positions into GNSS Skyplot, the direct transmission path of satellite signals blocked by the moving objects can be identified and excluded from positioning. Finally, GNSS positioning is estimated by the weighted least square (WLS) method based on the remaining satellites after the NLOS exclusion. Both static and dynamic experiments are conducted in Hong Kong. The results show that the proposed NLOS exclusion using LiDAR-based perception can greatly improve the GNSS single point positioning (SPP) performance.

Deep Learning for Generic Object Detection: A Survey Object detection, one of the most fundamental and challenging problems in computer vision, seeks to locate object instances from a large number of predefined categories in natural images. Deep learning techniques have emerged as a powerful strategy for learning feature representations directly from data and have led to remarkable breakthroughs in the field of generic object detection. Given this period of rapid evolution, the goal of this paper is to provide a comprehensive survey of the recent achievements in this field brought about by deep learning techniques. More than 300 research contributions are included in this survey, covering many aspects of generic object detection: detection frameworks, object feature representation, object proposal generation, context modeling, training strategies, and evaluation metrics. We finish the survey by identifying promising directions for future research.

Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations. Despite progress in perceptual tasks such as image classification, computers still perform poorly on cognitive tasks such as image description and question answering. Cognition is core to tasks that involve not just recognizing, but reasoning about our visual world. However, models used to tackle the rich content in images for cognitive tasks are still being trained using the same datasets designed for perceptual tasks. To achieve success at cognitive tasks, models need to understand the interactions and relationships between objects in an image. When asked \"What vehicle is the person riding?\", computers will need to identify the objects in an image as well as the relationships riding(man, carriage) and pulling(horse, carriage) to answer correctly that \"the person is riding a horse-drawn carriage.\" In this paper, we present the Visual Genome dataset to enable the modeling of such relationships. We collect dense annotations of objects, attributes, and relationships within each image to learn these models. Specifically, our dataset contains over 108K images where each image has an average of $$35$$35 objects, $$26$$26 attributes, and $$21$$21 pairwise relationships between objects. We canonicalize the objects, attributes, relationships, and noun phrases in region descriptions and questions answer pairs to WordNet synsets. Together, these annotations represent the densest and largest dataset of image descriptions, objects, attributes, relationships, and question answer pairs.

A standalone RFID Indoor Positioning System Using Passive Tags Indoor positioning systems (IPSs) locate objects in closed structures such as office buildings, hospitals, stores, factories, and warehouses, where Global Positioning System devices generally do not work. Most available systems apply wireless concepts, optical tracking, and/or ultrasound. This paper presents a standalone IPS using radio frequency identification (RFID) technology. The concept is ba...

The Labor of Fun: How Video Games Blur the Boundaries of Work and Play.

Picbreeder: evolving pictures collaboratively online Picbreeder is an online service that allows users to collaboratively evolve images. Like in other Interactive Evolutionary Computation (IEC) programs, users evolve images on Picbreeder by selecting ones that appeal to them to produce a new generation. However, Picbreeder also offers an online community in which to share these images, and most importantly, the ability to continue evolving others' images. Through this process of branching from other images, and through continually increasing image complexity made possible by the NeuroEvolution of Augmenting Topologies (NEAT) algorithm, evolved images proliferate unlike in any other current IEC systems. Participation requires no explicit talent from the users, thereby opening Picbreeder to the entire Internet community. This paper details how Picbreeder encourages innovation, featuring images that were collaboratively evolved.

Telecommunications Power Plant Damage Assessment for Hurricane Katrina– Site Survey and Follow-Up Results This paper extends knowledge of disaster impact on the telecommunications power infrastructure by discussing the effects of Hurricane Katrina based on an on-site survey conducted in October 2005 and on public sources. It includes observations about power infrastructure damage in wire-line and wireless networks. In general, the impact on centralized network elements was more severe than on the distributed portion of the grids. The main cause of outage was lack of power due to fuel supply disruptions, flooding and security issues. This work also describes the means used to restore telecommunications services and proposes ways to improve logistics, such as coordinating portable generator set deployment among different network operators and reducing genset fuel consumption by installing permanent photovoltaic systems at sites where long electric outages are likely. One long term solution is to use of distributed generation. It also discusses the consequences on telecom power technology and practices since the storm.

FADS: Circular/Spherical Sector based Forwarding Area Division and Adaptive Forwarding Area Selection routing protocol in WSNs. Without maintenance overhead make beaconless geographic routing protocol become an attractive routing scheme for wireless sensor networks (WSNs), however, without the knowledge of topology may result in high collision rate. In this paper, we firstly analyze the reason of CTS (Clear-To-Send) collision and classify it into Same-Slot collision and Distinct-Slot collision. Then, we proposed a novel routing protocol, Circular/Spherical Sector based Forwarding Area Division and Adaptive Forwarding Area Selection (FADS), which can avoid Distinct-Slot collision, reduce the probability of Same-Slot collision, and realize dynamic load balancing. 1) Forwarding area division ensures that every node within the same forwarding subarea is capable of hearing one another, thus avoiding Distinct-Slot collision. 2) Selecting forwarder from one of subarea decreases the number of contenders(candidates), hence the probability of Same-Slot collision is reduced. 3) Adaptive forwarding area selection dynamic channelizes the traffic to each subarea, thus realizing dynamic load balancing. Compared with related protocols, no matter in the 2D or 3D scenario, simulation results show the excellent performance of FADS in terms of packet delivery ratio, End-to-End latency, and energy consumption per packet, especially in dense networks and congested networks.

On Stability and Stabilization of T–S Fuzzy Systems With Time-Varying Delays via Quadratic Fuzzy Lyapunov Matrix This article proposes improved stability and stabilization criteria for Takagi–Sugeno (T–S) fuzzy systems with time-varying delays. First, a novel augmented fuzzy Lyapunov–Krasovskii functional (LKF) including the quadratic fuzzy Lyapunov matrix is constructed, which can provide much information of T–S fuzzy systems and help to achieve the lager allowable delay upper bounds. Then, improved delay-dependent stability and stabilization criteria are derived for the studied systems. Compared with the traditional methods, since the third-order Bessel–Legendre inequality and the extended reciprocally convex matrix inequality are well employed in the derivative of the constructed LKF to give tighter bounds of the single integral terms, the conservatism of derived criteria is further reduced. In addition, the quadratic fuzzy Lyapunov matrix introduced in LKF, which contains the quadratic membership functions, is also an important reason for obtaining less conservative results. Finally, numerical examples demonstrate that the proposed method is less conservative than some existing ones and the studied system can be well controlled by the designed controller.

Delay-Aware Dynamic Hypervisor Placement and Reconfiguration in Virtualized SDN Software defined networking (SDN) provides different functionality and resource sharing capabilities with the aid of virtualization. In virtualized SDN, multiple SDN tenants can bring their controllers and different functions in the same physical substrate. The SDN hypervisor provides the link between the physical network substrate and its SDN tenants. Distributed hypervisor architecture can handl...

Minimum interference routing of bandwidth guaranteed tunnels with MPLS traffic engineering applications This paper presents new algorithms for dynamic routing of bandwidth guaranteed tunnels, where tunnel routing requests arrive one by one and there is no a priori knowledge regarding future requests. This problem is motivated by the service provider needs for fast deployment of bandwidth guaranteed services. Offline routing algorithms cannot be used since they require a priori knowledge of all tunnel requests that are to be rooted. Instead, on-line algorithms that handle requests arriving one by one and that satisfy as many potential future demands as possible are needed. The newly developed algorithms are on-line algorithms and are based on the idea that a newly routed tunnel must follow a route that does not “interfere too much” with a route that may he critical to satisfy a future demand. We show that this problem is NP-hard. We then develop path selection heuristics which are based on the idea of deferred loading of certain “critical” links. These critical links are identified by the algorithm as links that, if heavily loaded, would make it impossible to satisfy future demands between certain ingress-egress pairs. Like min-hop routing, the presented algorithm uses link-state information and some auxiliary capacity information for path selection. Unlike previous algorithms, the proposed algorithm exploits any available knowledge of the network ingress-egress points of potential future demands, even though the demands themselves are unknown. If all nodes are ingress-egress nodes, the algorithm can still be used, particularly to reduce the rejection rate of requests between a specified subset of important ingress-egress pairs. The algorithm performs well in comparison to previously proposed algorithms on several metrics like the number of rejected demands and successful rerouting of demands upon link failure

The set cover with pairs problem We consider a generalization of the set cover problem, in which elements are covered by pairs of objects, and we are required to find a minimum cost subset of objects that induces a collection of pairs covering all elements. Formally, let U be a ground set of elements and let ${\cal S}$ be a set of objects, where each object i has a non-negative cost wi. For every $\{ i, j \} \subseteq {\cal S}$, let ${\cal C}(i,j)$ be the collection of elements in U covered by the pair { i, j }. The set cover with pairs problem asks to find a subset $A \subseteq {\cal S}$ such that $\bigcup_{ \{ i, j \} \subseteq A } {\cal C}(i,j) = U$ and such that ∑i∈Awi is minimized. In addition to studying this general problem, we are also concerned with developing polynomial time approximation algorithms for interesting special cases. The problems we consider in this framework arise in the context of domination in metric spaces and separation of point sets.

Control Plane Latency With SDN Network Hypervisors: The Cost of Virtualization. Software defined networking (SDN) network hypervisors provide the functionalities needed for virtualizing software-defined networks. Hypervisors sit logically between the multiple virtual SDN networks (vSDNs), which reside on the underlying physical SDN network infrastructure, and the corresponding tenant (vSDN) controllers. Different SDN network hypervisor architectures have mainly been explored through proof-of-concept implementations. We fundamentally advance SDN network hypervisor research by conducting a model-based analysis of SDN hypervisor architectures. Specifically, we introduce mixed integer programming formulations for four different SDN network hypervisor architectures. Our model formulations can also optimize the placement of multi-controller switches in virtualized OpenFlow-enabled SDN networks. We employ our models to quantitatively examine the optimal placement of the hypervisor instances. We compare the control plane latencies of the different SDN hypervisor architectures and quantify the cost of virtualization, i.e., the latency overhead due to virtualizing SDN networks via hypervisors. For generalization, we quantify how the hypervisor architectures behave for different network topologies. Our model formulations and the insights drawn from our evaluations inform network operators about the trade-offs of the different hypervisor architectures and help choosing an architecture according to operator demands.

Flow Setup Latency in SDN Networks. In software-defined networking, the typical switch-controller cycle, from generating a network event notification at the controller until the flow rules are installed at the switches, is not an instantaneous activity. Our measurement results show that this has serious implications on the performance of flow setup procedure, specifically for larger networks: we observe that, even with software swit...

A sub-constant error-probability low-degree test, and a sub-constant error-probability PCP characterization of NP

Completely derandomized self-adaptation in evolution strategies. This paper puts forward two useful methods for self-adaptation of the mutation distribution - the concepts of derandomization and cumulation. Principle shortcomings of the concept of mutative strategy parameter control and two levels of derandomization are reviewed. Basic demands on the self-adaptation of arbitrary (normal) mutation distributions are developed. Applying arbitrary, normal mutation distributions is equivalent to applying a general, linear problem encoding. The underlying objective of mutative strategy parameter control is roughly to favor previously selected mutation steps in the future. If this objective is pursued rigorously, a completely derandomized self-adaptation scheme results, which adapts arbitrary normal mutation distributions. This scheme, called covariance matrix adaptation (CMA), meets the previously stated demands. It can still be considerably improved by cumulation - utilizing an evolution path rather than single search steps. Simulations on various test functions reveal local and global search properties of the evolution strategy with and without covariance matrix adaptation. Their performances are comparable only on perfectly scaled functions. On badly scaled, non-separable functions usually a speed up factor of several orders of magnitude is observed. On moderately mis-scaled functions a speed up factor of three to ten can be expected.

A survey of socially interactive robots This paper reviews “socially interactive robots”: robots for which social human–robot interaction is important. We begin by discussing the context for socially interactive robots, emphasizing the relationship to other research fields and the different forms of “social robots”. We then present a taxonomy of design methods and system components used to build socially interactive robots. Finally, we describe the impact of these robots on humans and discuss open issues. An expanded version of this paper, which contains a survey and taxonomy of current applications, is available as a technical report [T. Fong, I. Nourbakhsh, K. Dautenhahn, A survey of socially interactive robots: concepts, design and applications, Technical Report No. CMU-RI-TR-02-29, Robotics Institute, Carnegie Mellon University, 2002].

Energy Efficiency Resource Allocation For D2d Communication Network Based On Relay Selection In order to solve the problem of spectrum resource shortage and energy consumption, we put forward a new model that combines with D2D communication and energy harvesting technology: energy harvesting-aided D2D communication network under the cognitive radio (EHA-CRD), where the D2D users harvest energy from the base station and the D2D source communicate with D2D destination by D2D relays. Our goals are to investigate the maximization energy efficiency (EE) of the network by joint time allocation and relay selection while taking into the constraints of the signal-to-noise ratio of D2D and the rates of the Cellular users. During this process, the energy collection time and communication time are randomly allocated. The maximization problem of EE can be divided into two sub-problems: (1) relay selection problem; (2) time optimization problem. For the first sub-problem, we propose a weighted sum maximum algorithm to select the best relay. For the last sub-problem, the EE maximization problem is non-convex problem with time. Thus, by using fractional programming theory, we transform it into a standard convex optimization problem, and we propose the optimization iterative algorithm to solve the convex optimization problem for obtaining the optimal solution. And, the simulation results show that the proposed relay selection algorithm and time optimization algorithm are significantly improved compared with the existing algorithms.

A communication robot in a shopping mall This paper reports our development of a communication robot for use in a shopping mall to provide shopping information, offer route guidance, and build rapport. In the development, the major difficulties included sensing human behaviors, conversation in a noisy daily environment, and the needs of unexpected miscellaneous knowledge in the conversation. We chose a networkrobot system approach, where a single robot's poor sensing capability and knowledge are supplemented by ubiquitous sensors and a human operator. The developed robot system detects a person with floor sensors to initiate interaction, identifies individuals with radio-frequency identification (RFID) tags, gives shopping information while chatting, and provides route guidance with deictic gestures. The robotwas partially teleoperated to avoid the difficulty of speech recognition as well as to furnish a new kind of knowledge that only humans can flexibly provide. The information supplied by a human operator was later used to increase the robot's autonomy. For 25 days in a shopping mall, we conducted a field trial and gathered 2642 interactions. A total of 235 participants signed up to use RFID tags and, later, provided questionnaire responses. The questionnaire results are promising in terms of the visitors' perceived acceptability as well as the encouragement of their shopping activities. The results of the teleoperation analysis revealed that the amount of teleoperation gradually decreased, which is also promising.

Minimum acceleration criterion with constraints implies bang-bang control as an underlying principle for optimal trajectories of arm reaching movements. Rapid arm-reaching movements serve as an excellent test bed for any theory about trajectory formation. How are these movements planned? A minimum acceleration criterion has been examined in the past, and the solution obtained, based on the Euler-Poisson equation, failed to predict that the hand would begin and end the movement at rest (i.e., with zero acceleration). Therefore, this criterion was rejected in favor of the minimum jerk, which was proved to be successful in describing many features of human movements. This letter follows an alternative approach and solves the minimum acceleration problem with constraints using Pontryagin's minimum principle. We use the minimum principle to obtain minimum acceleration trajectories and use the jerk as a control signal. In order to find a solution that does not include nonphysiological impulse functions, constraints on the maximum and minimum jerk values are assumed. The analytical solution provides a three-phase piecewise constant jerk signal (bang-bang control) where the magnitude of the jerk and the two switching times depend on the magnitude of the maximum and minimum available jerk values. This result fits the observed trajectories of reaching movements and takes into account both the extrinsic coordinates and the muscle limitations in a single framework. The minimum acceleration with constraints principle is discussed as a unifying approach for many observations about the neural control of movements.

An Automatic Screening Approach for Obstructive Sleep Apnea Diagnosis Based on Single-Lead Electrocardiogram Traditional approaches for obstructive sleep apnea (OSA) diagnosis are apt to using multiple channels of physiological signals to detect apnea events by dividing the signals into equal-length segments, which may lead to incorrect apnea event detection and weaken the performance of OSA diagnosis. This paper proposes an automatic-segmentation-based screening approach with the single channel of Electrocardiogram (ECG) signal for OSA subject diagnosis, and the main work of the proposed approach lies in three aspects: (i) an automatic signal segmentation algorithm is adopted for signal segmentation instead of the equal-length segmentation rule; (ii) a local median filter is improved for reduction of the unexpected RR intervals before signal segmentation; (iii) the designed OSA severity index and additional admission information of OSA suspects are plugged into support vector machine (SVM) for OSA subject diagnosis. A real clinical example from PhysioNet database is provided to validate the proposed approach and an average accuracy of 97.41% for subject diagnosis is obtained which demonstrates the effectiveness for OSA diagnosis.

A robust medical image watermarking against salt and pepper noise for brain MRI images. The ever-growing numbers of medical digital images and the need to share them among specialists and hospitals for better and more accurate diagnosis require that patients' privacy be protected. During the transmission of medical images between hospitals or specialists through the network, the main priority is to protect a patient's documents against any act of tampering by unauthorised individuals. Because of this, there is a need for medical image authentication scheme to enable proper diagnosis on patient. In addition, medical images are also susceptible to salt and pepper impulse noise through the transmission in communication channels. This noise may also be intentionally used by the invaders to corrupt the embedded watermarks inside the medical images. A common drawback of existing watermarking methods is their weakness against salt and pepper noise. The research carried out in this work addresses the issue of designing a new watermarking method that can withstand high density of salt and pepper noise for brain MRI images. For this purpose, combination of a spatial domain watermarking method, channel coding and noise filtering schemes are used. The region of non-interest (RONI) of MRI images from five different databases are used as embedding area and electronic patient record (EPR) is considered as embedded data. The quality of watermarked image is evaluated using Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM), and the accuracy of the extracted watermark is assessed in terms of Bit Error Rate (BER).

Convert Harm Into Benefit: A Coordination-Learning Based Dynamic Spectrum Anti-Jamming Approach This paper mainly investigates the multi-user anti-jamming spectrum access problem. Using the idea of “converting harm into benefit,” the malicious jamming signals projected by the enemy are utilized by the users as the coordination signals to guide spectrum coordination. An “internal coordination-external confrontation” multi-user anti-jamming access game model is constructed, and the existence of Nash equilibrium (NE) as well as correlated equilibrium (CE) is demonstrated. A coordination-learning based anti-jamming spectrum access algorithm (CLASA) is designed to achieve the CE of the game. Simulation results show the convergence, and effectiveness of the proposed CLASA algorithm, and indicate that our approach can help users confront the malicious jammer, and coordinate internal spectrum access simultaneously without information exchange. Last but not least, the fairness of the proposed approach under different jamming attack patterns is analyzed, which illustrates that this approach provides fair anti-jamming spectrum access opportunities under complicated jamming pattern.

Deep Learning for Understanding Faces: Machines May Be Just as Good, or Better, than Humans. Recent developments in deep convolutional neural networks (DCNNs) have shown impressive performance improvements on various object detection/recognition problems. This has been made possible due to the availability of large annotated data and a better understanding of the nonlinear mapping between images and class labels, as well as the affordability of powerful graphics processing units (GPUs). T...

A survey on deep learning based face recognition. Deep learning, in particular the deep convolutional neural networks, has received increasing interests in face recognition recently, and a number of deep learning methods have been proposed. This paper summarizes about 330 contributions in this area. It reviews major deep learning concepts pertinent to face image analysis and face recognition, and provides a concise overview of studies on specific face recognition problems, such as handling variations in pose, age, illumination, expression, and heterogeneous face matching. A summary of databases used for deep face recognition is given as well. Finally, some open challenges and directions are discussed for future research.

The Effect of Broad and Specific Demographic Homogeneity on the Imposter Distributions and False Match Rates in Face Recognition Algorithm Performance The growing adoption of biometric identity systems, notably face recognition, has raised questions regarding whether performance is equitable across demographic groups. Prior work on this issue showed that performance of face recognition systems varies with demographic variables. However, biometric systems make two distinct types of matching errors, which lead to different outcomes for users depending on the technology use case. In this research, we develop a framework for classifying biometric performance differentials that separately considers the effect of false positive and false negative outcomes, and show that oft-cited evidence regarding biometric equitability has focused on primarily on false-negatives. We then correlate demographic variables with false-positive outcomes in a diverse population using a commercial face recognition algorithm, and show that false match rate (FMR) at a fixed threshold increases >400-fold for broadly homogeneous groups (individuals of the same age, same gender, and same race) relative to heterogeneous groups. This was driven by systematic shifts in the tails of the imposter distribution impelled primarily by homogeneity in race and gender. For specific demographic groups, we observed the highest false match rate for older males that self-identified as White and the lowest for older males that self-identified as Black or African American. The magnitude of FMR differentials between specific homogeneous groups (<;3-fold) was modest in comparison with the FMR increase associated with broad demographic homogeneity. These results demonstrate the false positive outcomes of face recognition systems are not simply linked to single demographic factors, and that a careful consideration of interactions between multiple factors is needed when considering the equitability of these systems.

The FERET Evaluation Methodology for Face-Recognition Algorithms Two of the most critical requirements in support of producing reliable face-recognition systems are a large database of facial images and a testing procedure to evaluate systems. The Face Recognition Technology (FERET) program has addressed both issues through the FERET database of facial images and the establishment of the FERET tests. To date, 14,126 images from 1,199 individuals are included in the FERET database, which is divided into development and sequestered portions of the database. In September 1996, the FERET program administered the third in a series of FERET face-recognition tests. The primary objectives of the third test were to 1) assess the state of the art, 2) identify future areas of research, and 3) measure algorithm performance.

On-road prediction of driver's intent with multimodal sensory cues By predicting a driver's maneuvers before they occur, a driver-assistance system can prepare for or avoid dangerous situations. This article describes a real-time, on-road lane-change-intent detector that can enhance driver safety.

Real-time system for monitoring driver vigilance This paper presents a nonintrusive prototype computer vision system for monitoring a driver's vigilance in real time. It is based on a hardware system for the real-time acquisition of a driver's images using an active IR illuminator and the software implementation for monitoring some visual behaviors that characterize a driver's level of vigilance. Six parameters are calculated: Percent eye closure (PERCLOS), eye closure duration, blink frequency, nodding frequency, face position, and fixed gaze. These parameters are combined using a fuzzy classifier to infer the level of inattentiveness of the driver. The use of multiple visual parameters and the fusion of these parameters yield a more robust and accurate inattention characterization than by using a single parameter. The system has been tested with different sequences recorded in night and day driving conditions in a motorway and with different users. Some experimental results and conclusions about the performance of the system are presented

Modelling and Control Strategies in Path Tracking Control for Autonomous Ground Vehicles: A Review of State of the Art and Challenges. Autonomous vehicle field of study has seen considerable researches within three decades. In the last decade particularly, interests in this field has undergone tremendous improvement. One of the main aspects in autonomous vehicle is the path tracking control, focusing on the vehicle control in lateral and longitudinal direction in order to follow a specified path or trajectory. In this paper, path tracking control is reviewed in terms of the basic vehicle model usually used; the control strategies usually employed in path tracking control, and the performance criteria used to evaluate the controller's performance. Vehicle model is categorised into several types depending on its linearity and the type of behaviour it simulates, while path tracking control is categorised depending on its approach. This paper provides critical review of each of these aspects in terms of its usage and disadvantages/advantages. Each aspect is summarised for better overall understanding. Based on the critical reviews, main challenges in the field of path tracking control is identified and future research direction is proposed. Several promising advancement is proposed with the main prospect is focused on adaptive geometric controller developed on a nonlinear vehicle model and tested with hardware-in-the-loop (HIL). It is hoped that this review can be treated as preliminary insight into the choice of controllers in path tracking control development for an autonomous ground vehicle.

Data-Driven Estimation of Driver Attention Using Calibration-Free Eye Gaze and Scene Features Driver attention estimation is one of the key technologies for intelligent vehicles. The existing related methods only focus on the scene image or the driver&#39;s gaze or head pose. The purpose of this article is to propose a more reasonable and feasible method based on a dual-view scene with calibration-free gaze direction. According to human visual mechanisms, the low-level features, static visual ...

An introduction to ROC analysis Receiver operating characteristics (ROC) graphs are useful for organizing classifiers and visualizing their performance. ROC graphs are commonly used in medical decision making, and in recent years have been used increasingly in machine learning and data mining research. Although ROC graphs are apparently simple, there are some common misconceptions and pitfalls when using them in practice. The purpose of this article is to serve as an introduction to ROC graphs and as a guide for using them in research.

Precomputing Oblivious Transfer Alice and Bob are too untrusting of computer scientists to let their privacy depend on unproven assumptions such as the existence of one-way functions. Firm believers in Schrödinger and Heisenberg, they might accept a quantum OT device, but IBM’s prototype is not yet portable. Instead, as part of their prenuptial agreement, they decide to visit IBM and perform some OT’s in advance, so that any later divorces, coin-flipping or other important interactions can be done more conveniently, without needing expensive third parties. Unfortunately, OT can’t be done in advance in a direct way, because even though Bob might not know what bit Alice will later send (even if she first sends a random bit and later corrects it, for example), he would already know which bit or bits he will receive. We address the problem of precomputing oblivious transfer and show that OT can be precomputed at a cost of Θ(κ) prior transfers (a tight bound). In contrast, we show that variants of OT, such as one-out-of-two OT, can be precomputed using only one prior transfer. Finally, we show that all variants can be reduced to a single precomputed one-out-of-two oblivious transfer.

CellSense: A Probabilistic RSSI-Based GSM Positioning System Context-aware applications have been gaining huge interest in the last few years. With cell phones becoming ubiquitous computing devices, cell phone localization has become an important research problem. In this paper, we present CellSense, a probabilistic RSSI-based fingerprinting location determination system for GSM phones. We discuss the challenges of implementing a probabilistic fingerprinting localization technique in GSM networks and present the details of the CellSense system and how it addresses the challenges. To evaluate our proposed system, we implemented CellSense on Android-based phones. Results for two different testbeds, representing urban and rural environments, show that CellSense provides at least 23.8% enhancement in accuracy in rural areas and at least 86.4% in urban areas compared to other RSSI-based GSM localization systems. This comes with a minimal increase in computational requirements. We also evaluate the effect of changing the different system parameters on the accuracy-complexity tradeoff.

Image Feature Extraction in Encrypted Domain With Privacy-Preserving SIFT Privacy has received considerable attention but is still largely ignored in the multimedia community. Consider a cloud computing scenario where the server is resource-abundant, and is capable of finishing the designated tasks. It is envisioned that secure media applications with privacy preservation will be treated seriously. In view of the fact that scale-invariant feature transform (SIFT) has been widely adopted in various fields, this paper is the first to target the importance of privacy-preserving SIFT (PPSIFT) and to address the problem of secure SIFT feature extraction and representation in the encrypted domain. As all of the operations in SIFT must be moved to the encrypted domain, we propose a privacy-preserving realization of the SIFT method based on homomorphic encryption. We show through the security analysis based on the discrete logarithm problem and RSA that PPSIFT is secure against ciphertext only attack and known plaintext attack. Experimental results obtained from different case studies demonstrate that the proposed homomorphic encryption-based privacy-preserving SIFT performs comparably to the original SIFT and that our method is useful in SIFT-based privacy-preserving applications.

Generating Natural Language Adversarial Examples. Deep neural networks (DNNs) are vulnerable to adversarial examples, perturbations to correctly classified examples which can cause the model to misclassify. In the image domain, these perturbations are often virtually indistinguishable to human perception, causing humans and state-of-the-art models to disagree. However, in the natural language domain, small perturbations are clearly perceptible, and the replacement of a single word can drastically alter the semantics of the document. Given these challenges, we use a black-box population-based optimization algorithm to generate semantically and syntactically similar adversarial examples that fool well-trained sentiment analysis and textual entailment models with success rates of 97% and 70%, respectively. We additionally demonstrate that 92.3% of the successful sentiment analysis adversarial examples are classified to their original label by 20 human annotators, and that the examples are perceptibly quite similar. Finally, we discuss an attempt to use adversarial training as a defense, but fail to yield improvement, demonstrating the strength and diversity of our adversarial examples. We hope our findings encourage researchers to pursue improving the robustness of DNNs in the natural language domain.

Design and Validation of a Cable-Driven Asymmetric Back Exosuit Lumbar spine injuries caused by repetitive lifting rank as the most prevalent workplace injury in the United States. While these injuries are caused by both symmetric and asymmetric lifting, asymmetric is often more damaging. Many back devices do not address asymmetry, so we present a new system called the Asymmetric Back Exosuit (ABX). The ABX addresses this important gap through unique design geometry and active cable-driven actuation. The suit allows the user to move in a wide range of lumbar trajectories while the “X” pattern cable routing allows variable assistance application for these trajectories. We also conducted a biomechanical analysis in OpenSim to map assistive cable force to effective lumbar torque assistance for a given trajectory, allowing for intuitive controller design in the lumbar joint space over the complex kinematic chain for varying lifting techniques. Human subject experiments illustrated that the ABX reduced lumbar erector spinae muscle activation during symmetric and asymmetric lifting by an average of 37.8% and 16.0%, respectively, compared to lifting without the exosuit. This result indicates the potential for our device to reduce lumbar injury risk.

Addictive links: engaging students through adaptive navigation support and open social student modeling Empirical studies of adaptive annotation in the educational context have demonstrated that it can help students to acquire knowledge faster, improve learning outcomes, reduce navigational overhead, and encourage non-sequential navigation. Over the last 8 years we have explored a lesser known effect of adaptive annotation -- its ability to significantly increase student engagement in working with non-mandatory educational content. In the presence of adaptive link annotation, students tend to access significantly more learning content; they stay with it longer, return to it more often and explore a wider variety of learning resources. This talk will present an overview of our exploration of the addictive links effect in many course-long studies, which we ran in several domains (C, SQL and Java programming), for several types of learning content (quizzes, problems, interactive examples). The first part of the talk will review our exploration of a more traditional knowledge-based personalization approach and the second part will focus on more recent studies of social navigation and open social student modeling.

The Sybil Attack Large-scale peer-to-peer systems facesecurity threats from faulty or hostile remotecomputing elements. To resist these threats, manysuch systems employ redundancy. However, if asingle faulty entity can present multiple identities,it can control a substantial fraction of the system,thereby undermining this redundancy. Oneapproach to preventing these &quot;Sybil attacks&quot; is tohave a trusted agency certify identities. Thispaper shows that, without a logically centralizedauthority, Sybil...

BLEU: a method for automatic evaluation of machine translation Human evaluations of machine translation are extensive but expensive. Human evaluations can take months to finish and involve human labor that can not be reused. We propose a method of automatic machine translation evaluation that is quick, inexpensive, and language-independent, that correlates highly with human evaluation, and that has little marginal cost per run. We present this method as an automated understudy to skilled human judges which substitutes for them when there is need for quick or frequent evaluations.

Computational thinking Summary form only given. My vision for the 21st century, Computational Thinking, will be a fundamental skill used by everyone in the world. To reading, writing, and arithmetic, we should add computational thinking to every child's analytical ability. Computational thinking involves solving problems, designing systems, and understanding human behavior by drawing on the concepts fundamental to computer science. Thinking like a computer scientist means more than being able to program a computer. It requires the ability to abstract and thus to think at multiple levels of abstraction. In this talk I will give many examples of computational thinking, argue that it has already influenced other disciplines, and promote the idea that teaching computational thinking can not only inspire future generations to enter the field of computer science but benefit people in all fields.

Fuzzy logic in control systems: fuzzy logic controller. I.

Switching between stabilizing controllers This paper deals with the problem of switching between several linear time-invariant (LTI) controllers—all of them capable of stabilizing a speci4c LTI process—in such a way that the stability of the closed-loop system is guaranteed for any switching sequence. We show that it is possible to 4nd realizations for any given family of controller transfer matrices so that the closed-loop system remains stable, no matter how we switch among the controller. The motivation for this problem is the control of complex systems where con8icting requirements make a single LTI controller unsuitable. ? 2002 Published by Elsevier Science Ltd.

Tabu Search - Part I

Bidirectional recurrent neural networks In the first part of this paper, a regular recurrent neural network (RNN) is extended to a bidirectional recurrent neural network (BRNN). The BRNN can be trained without the limitation of using input information just up to a preset future frame. This is accomplished by training it simultaneously in positive and negative time direction. Structure and training procedure of the proposed network are explained. In regression and classification experiments on artificial data, the proposed structure gives better results than other approaches. For real data, classification experiments for phonemes from the TIMIT database show the same tendency. In the second part of this paper, it is shown how the proposed bidirectional structure can be easily modified to allow efficient estimation of the conditional posterior probability of complete symbol sequences without making any explicit assumption about the shape of the distribution. For this part, experiments on real data are reported

An intensive survey of fair non-repudiation protocols With the phenomenal growth of the Internet and open networks in general, security services, such as non-repudiation, become crucial to many applications. Non-repudiation services must ensure that when Alice sends some information to Bob over a network, neither Alice nor Bob can deny having participated in a part or the whole of this communication. Therefore a fair non-repudiation protocol has to generate non-repudiation of origin evidences intended to Bob, and non-repudiation of receipt evidences destined to Alice. In this paper, we clearly define the properties a fair non-repudiation protocol must respect, and give a survey of the most important non-repudiation protocols without and with trusted third party (TTP). For the later ones we discuss the evolution of the TTP's involvement and, between others, describe the most recent protocol using a transparent TTP. We also discuss some ad-hoc problems related to the management of non-repudiation evidences.

Dynamic movement and positioning of embodied agents in multiparty conversations For embodied agents to engage in realistic multiparty conversation, they must stand in appropriate places with respect to other agents and the environment. When these factors change, such as an agent joining the conversation, the agents must dynamically move to a new location and/or orientation to accommodate. This paper presents an algorithm for simulating movement of agents based on observed human behavior using techniques developed for pedestrian movement in crowd simulations. We extend a previous group conversation simulation to include an agent motion algorithm. We examine several test cases and show how the simulation generates results that mirror real-life conversation settings.

An improved genetic algorithm with conditional genetic operators and its application to set-covering problem The genetic algorithm (GA) is a popular, biologically inspired optimization method. However, in the GA there is no rule of thumb to design the GA operators and select GA parameters. Instead, trial-and-error has to be applied. In this paper we present an improved genetic algorithm in which crossover and mutation are performed conditionally instead of probability. Because there are no crossover rate and mutation rate to be selected, the proposed improved GA can be more easily applied to a problem than the conventional genetic algorithms. The proposed improved genetic algorithm is applied to solve the set-covering problem. Experimental studies show that the improved GA produces better results over the conventional one and other methods.

Automated Demand Response From Home Energy Management System Under Dynamic Pricing and Power and Comfort Constraints This paper presents a comprehensive and general optimization-based home energy management controller, incorporating several classes of domestic appliances including deferrable, curtailable, thermal, and critical ones. The operations of the appliances are controlled in response to dynamic price signals to reduce the consumer's electricity bill whilst minimizing the daily volume of curtailed energy, and therefore considering the user's comfort level. To avoid shifting a large portion of consumer demand toward the least price intervals, which could create network issues due to loss of diversity, higher prices are applied when the consumer's demand goes beyond a prescribed power threshold. The arising mixed integer nonlinear optimization problem is solved in an iterative manner rolling throughout the day to follow the changes in the anticipated price signals and the variations in the controller inputs while information is updated. The results from different realistic case studies show the effectiveness of the proposed controller in minimizing the household's daily electricity bill while {preserving} comfort level, as well as preventing creation of new least-price peaks.

Scalable and Privacy-Preserving Data Sharing Based on Blockchain. With the development of network technology and cloud computing, data sharing is becoming increasingly popular, and many scholars have conducted in-depth research to promote its flourish. As the scale of data sharing expands, its privacy protection has become a hot issue in research. Moreover, in data sharing, the data is usually maintained in multiple parties, which brings new challenges to protect the privacy of these multi-party data. In this paper, we propose a trusted data sharing scheme using blockchain. We use blockchain to prevent the shared data from being tampered, and use the Paillier cryptosystem to realize the confidentiality of the shared data. In the proposed scheme, the shared data can be traded, and the transaction information is protected by using the (p, t)-threshold Paillier cryptosystem. We conduct experiments in cloud storage scenarios and the experimental results demonstrate the efficiency and effectiveness of the proposed scheme.

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

Resilience-Driven Quantitative Analysis of Vehicle Platooning Service Vehicle platooning can be applied to cooperative downloading and uploading (CDU) services through the cooperation between lead vehicle and non-lead vehicles. CDU service can be completed cooperatively by containers constructed in vehicles of the vehicle platooning system. Containers in vehicles may suffer from potential attacks which can lead to resource degradation and further cause the decrease in the dependability and security of CDU service. Failover technique can be deployed in an in-vehicle system and then performs protective operation for mitigating the impact of resources degradation and guaranteeing service resilience. This paper aims to quantitatively analyze the availability, reliability and security of CDU service in the vehicle platooning system deploying failover technique. We construct an analytical model to describe the interaction between resources degradation and rejuvenation behaviors of the vehicle platooning system. Then we derive the formulas of availability, Mean Time to Failure (MTTF) for evaluating reliability and Mean Security Capacity to Failure (MSCF) for evaluating security. Simulation and analytical experiments are applied to demonstrate the approximate accuracy of our model and investigate the impact of system parameters on metrics, respectively, which helps service providers make decisions on how to organize vehicles to perform CDU service and how to set system parameter values for maximizing their benefits.

Analysis of Software Aging in a Web Server Several recent studies have reported & examined the phenomenon that long-running software systems show an increasing failure rate and/or a progressive degradation of their performance. Causes of this phenomenon, which has been referred to as &#34;software aging&#34;, are the accumulation of internal error conditions, and the depletion of operating system resources. A proactive technique called &#34;software r...

Container Network Functions: Bringing NFV to the Network Edge. In order to cope with the increasing network utilization driven by new mobile clients, and to satisfy demand for new network services and performance guarantees, telecommunication service providers are exploiting virtualization over their network by implementing network services in virtual machines, decoupled from legacy hardware accelerated appliances. This effort, known as NFV, reduces OPEX and ...

Reliability-Aware Service Chaining In Carrier-Grade Softwarized Networks. Network Function Virtualization (NFV) has revolutionized service provisioning in cloud datacenter networks. It enables the complete decoupling of Network Functions (NFs) from the physical hardware middle boxes that network operators deploy for implementing service-specific and strictly ordered NF chains. Precisely, NFV allows for dispatching NFs as instances of plain software called virtual networ...

Analyzing Software Rejuvenation Techniques in a Virtualized System: Service Provider and User Views Virtualization technology has promoted the fast development and deployment of cloud computing, and is now becoming an enabler of Internet of Everything. Virtual machine monitor (VMM), playing a critical role in a virtualized system, is software and hence it suffers from software aging after a long continuous running as well as software crashes due to elusive faults. Software rejuvenation techniques can be adopted to reduce the impact of software aging. Although there existed analytical model-based approaches for evaluating software rejuvenation techniques, none analyzed both application service (AS) availability and job completion time in a virtualized system with live virtual machine (VM) migration. This paper aims to quantitatively analyze software rejuvenation techniques from service provider and user views in a virtualized system deploying VMM reboot and live VM migration techniques for rejuvenation, under the condition that all the aging time, failure time, VMM fixing time and live VM migration time follow general distributions. We construct an analytical model by using a semi-Markov process (SMP) and derive formulas for calculating AS availability and job completion time. By analytical experiments, we can obtain the optimal migration trigger intervals for achieving the approximate maximum AS availability and the approximate minimum job completion time, and then service providers can make decisions for maximizing the benefits of service providers and users by adjusting parameter values.

BLEU: a method for automatic evaluation of machine translation Human evaluations of machine translation are extensive but expensive. Human evaluations can take months to finish and involve human labor that can not be reused. We propose a method of automatic machine translation evaluation that is quick, inexpensive, and language-independent, that correlates highly with human evaluation, and that has little marginal cost per run. We present this method as an automated understudy to skilled human judges which substitutes for them when there is need for quick or frequent evaluations.

Massive MIMO for next generation wireless systems Multi-user MIMO offers big advantages over conventional point-to-point MIMO: it works with cheap single-antenna terminals, a rich scattering environment is not required, and resource allocation is simplified because every active terminal utilizes all of the time-frequency bins. However, multi-user MIMO, as originally envisioned, with roughly equal numbers of service antennas and terminals and frequency-division duplex operation, is not a scalable technology. Massive MIMO (also known as large-scale antenna systems, very large MIMO, hyper MIMO, full-dimension MIMO, and ARGOS) makes a clean break with current practice through the use of a large excess of service antennas over active terminals and time-division duplex operation. Extra antennas help by focusing energy into ever smaller regions of space to bring huge improvements in throughput and radiated energy efficiency. Other benefits of massive MIMO include extensive use of inexpensive low-power components, reduced latency, simplification of the MAC layer, and robustness against intentional jamming. The anticipated throughput depends on the propagation environment providing asymptotically orthogonal channels to the terminals, but so far experiments have not disclosed any limitations in this regard. While massive MIMO renders many traditional research problems irrelevant, it uncovers entirely new problems that urgently need attention: the challenge of making many low-cost low-precision components that work effectively together, acquisition and synchronization for newly joined terminals, the exploitation of extra degrees of freedom provided by the excess of service antennas, reducing internal power consumption to achieve total energy efficiency reductions, and finding new deployment scenarios. This article presents an overview of the massive MIMO concept and contemporary research on the topic.

Deep Residual Learning for Image Recognition Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously. We explicitly reformulate the layers as learning residual functions with reference to the layer inputs, instead of learning unreferenced functions. We provide comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth. On the ImageNet dataset we evaluate residual nets with a depth of up to 152 layers - 8× deeper than VGG nets [40] but still having lower complexity. An ensemble of these residual nets achieves 3.57% error on the ImageNet test set. This result won the 1st place on the ILSVRC 2015 classification task. We also present analysis on CIFAR-10 with 100 and 1000 layers. The depth of representations is of central importance for many visual recognition tasks. Solely due to our extremely deep representations, we obtain a 28% relative improvement on the COCO object detection dataset. Deep residual nets are foundations of our submissions to ILSVRC & COCO 2015 competitions1, where we also won the 1st places on the tasks of ImageNet detection, ImageNet localization, COCO detection, and COCO segmentation.

Communication theory of secrecy systems THE problems of cryptography and secrecy systems furnish an interesting application of communication theory.1 In this paper a theory of secrecy systems is developed. The approach is on a theoretical level and is intended to complement the treatment found in standard works on cryptography.2 There, a detailed study is made of the many standard types of codes and ciphers, and of the ways of breaking them. We will be more concerned with the general mathematical structure and properties of secrecy systems.

A study on the use of non-parametric tests for analyzing the evolutionary algorithms' behaviour: a case study on the CEC'2005 Special Session on Real Parameter Optimization In recent years, there has been a growing interest for the experimental analysis in the field of evolutionary algorithms. It is noticeable due to the existence of numerous papers which analyze and propose different types of problems, such as the basis for experimental comparisons of algorithms, proposals of different methodologies in comparison or proposals of use of different statistical techniques in algorithms’ comparison.In this paper, we focus our study on the use of statistical techniques in the analysis of evolutionary algorithms’ behaviour over optimization problems. A study about the required conditions for statistical analysis of the results is presented by using some models of evolutionary algorithms for real-coding optimization. This study is conducted in two ways: single-problem analysis and multiple-problem analysis. The results obtained state that a parametric statistical analysis could not be appropriate specially when we deal with multiple-problem results. In multiple-problem analysis, we propose the use of non-parametric statistical tests given that they are less restrictive than parametric ones and they can be used over small size samples of results. As a case study, we analyze the published results for the algorithms presented in the CEC’2005 Special Session on Real Parameter Optimization by using non-parametric test procedures.

Implementing Vehicle Routing Algorithms

Switching Stabilization for a Class of Slowly Switched Systems In this technical note, the problem of switching stabilization for slowly switched linear systems is investigated. In particular, the considered systems can be composed of all unstable subsystems. Based on the invariant subspace theory, the switching signal with mode-dependent average dwell time (MDADT) property is designed to exponentially stabilize the underlying system. Furthermore, sufficient condition of stabilization for switched systems with all stable subsystems under MDADT switching is also given. The correctness and effectiveness of the proposed approaches are illustrated by a numerical example.

Neural network adaptive tracking control for a class of uncertain switched nonlinear systems. •Study the method of the tracking control of the switched uncertain nonlinear systems under arbitrary switching signal controller.•A multilayer neural network adaptive controller with multilayer weight norm adaptive estimation is been designed.•The adaptive law is expand from calculation the second layer weight of neural network to both of the two layers weight.•The controller proposed improve the tracking error performance of the closed-loop system greatly.

Energy harvesting algorithm considering max flow problem in wireless sensor networks. In Wireless Sensor Networks (WSNs), sensor nodes with poor energy always have bad effect on the data rate or max flow. These nodes are called bottleneck nodes. In this paper, in order to increase the max flow, we assume an energy harvesting WSNs environment to investigate the cooperation of multiple Mobile Chargers (MCs). MCs are mobile robots that use wireless charging technology to charge sensor nodes in WSNs. This means that in energy harvesting WSNs environments, sensor nodes can obtain energy replenishment by using MCs or collecting energy from nature by themselves. In our research, we use MCs to improve the energy of the sensor nodes by performing multiple rounds of unified scheduling, and finally achieve the purpose of increasing the max flow at sinks. Firstly, we model this problem as a Linear Programming (LP) to search the max flow in a round of charging scheduling and prove that the problem is NP-hard. In order to solve the problem, we propose a heuristic approach: deploying MCs in units of paths with the lowest energy node priority. To reduce the energy consumption of MCs and increase the charging efficiency, we also take the optimization of MCs’ moving distance into our consideration. Finally, we extend the method to multiple rounds of scheduling called BottleNeck. Simulation results show that Bottleneck performs well at increasing max flow.

Minimum interference routing of bandwidth guaranteed tunnels with MPLS traffic engineering applications This paper presents new algorithms for dynamic routing of bandwidth guaranteed tunnels, where tunnel routing requests arrive one by one and there is no a priori knowledge regarding future requests. This problem is motivated by the service provider needs for fast deployment of bandwidth guaranteed services. Offline routing algorithms cannot be used since they require a priori knowledge of all tunnel requests that are to be rooted. Instead, on-line algorithms that handle requests arriving one by one and that satisfy as many potential future demands as possible are needed. The newly developed algorithms are on-line algorithms and are based on the idea that a newly routed tunnel must follow a route that does not “interfere too much” with a route that may he critical to satisfy a future demand. We show that this problem is NP-hard. We then develop path selection heuristics which are based on the idea of deferred loading of certain “critical” links. These critical links are identified by the algorithm as links that, if heavily loaded, would make it impossible to satisfy future demands between certain ingress-egress pairs. Like min-hop routing, the presented algorithm uses link-state information and some auxiliary capacity information for path selection. Unlike previous algorithms, the proposed algorithm exploits any available knowledge of the network ingress-egress points of potential future demands, even though the demands themselves are unknown. If all nodes are ingress-egress nodes, the algorithm can still be used, particularly to reduce the rejection rate of requests between a specified subset of important ingress-egress pairs. The algorithm performs well in comparison to previously proposed algorithms on several metrics like the number of rejected demands and successful rerouting of demands upon link failure

The set cover with pairs problem We consider a generalization of the set cover problem, in which elements are covered by pairs of objects, and we are required to find a minimum cost subset of objects that induces a collection of pairs covering all elements. Formally, let U be a ground set of elements and let ${\cal S}$ be a set of objects, where each object i has a non-negative cost wi. For every $\{ i, j \} \subseteq {\cal S}$, let ${\cal C}(i,j)$ be the collection of elements in U covered by the pair { i, j }. The set cover with pairs problem asks to find a subset $A \subseteq {\cal S}$ such that $\bigcup_{ \{ i, j \} \subseteq A } {\cal C}(i,j) = U$ and such that ∑i∈Awi is minimized. In addition to studying this general problem, we are also concerned with developing polynomial time approximation algorithms for interesting special cases. The problems we consider in this framework arise in the context of domination in metric spaces and separation of point sets.

Pareto-optimal resilient controller placement in SDN-based core networks With the introduction of Software Defined Networking (SDN), the concept of an external and optionally centralized network control plane, i.e. controller, is drawing the attention of researchers and industry. A particularly important task in the SDN context is the placement of such external resources in the network. In this paper, we discuss important aspects of the controller placement problem with a focus on SDN-based core networks, including different types of resilience and failure tolerance. When several performance and resilience metrics are considered, there is usually no single best controller placement solution, but a trade-off between these metrics. We introduce our framework for resilient Pareto-based Optimal COntroller-placement (POCO) that provides the operator of a network with all Pareto-optimal placements. The ideas and mechanisms are illustrated using the Internet2 OS3E topology and further evaluated on more than 140 topologies of the Topology Zoo. In particular, our findings reveal that for most of the topologies more than 20% of all nodes need to be controllers to assure a continuous connection of all nodes to one of the controllers in any arbitrary double link or node failure scenario.

Renaissance: A Self-Stabilizing Distributed SDN Control Plane By introducing programmability, automated verification, and innovative debugging tools, Software-Defined Networks (SDNs) are poised to meet the increasingly stringent dependability requirements of today's communication networks. However, the design of fault-tolerant SDNs remains an open challenge. This paper considers the design of dependable SDNs through the lenses of self-stabilization - a very strong notion of fault-tolerance. In particular, we develop algorithms for an in-band and distributed control plane for SDNs, called Renaissance, which tolerates a wide range of (concurrent) controller, link, and communication failures. Our self-stabilizing algorithms ensure that after the occurrence of an arbitrary combination of failures, (i) every non-faulty SDN controller can eventually reach any switch in the network within a bounded communication delay (in the presence of a bounded number of concurrent failures) and (ii) every switch is managed by at least one non-faulty controller. We evaluate Renaissance through a rigorous worst-case analysis as well as a prototype implementation (based on OVS and Floodlight), and we report on our experiments using Mininet.

Enumerating Connected Induced Subgraphs: Improved Delay And Experimental Comparison We consider the problem of enumerating all connected induced subgraphs of order k in an undirected graph G = (V, E). Our main results are two enumeration algorithms with a delay of O(k(2)Delta) where Delta is the maximum degree in the input graph. This improves upon a previous delay bound (Elbassioni, 2015) for this problem. Moreover, we show that these two algorithms can be adapted to give algorithms for the problem of enumerating all connected induced subgraphs of order at most k with a delay of O(k + Delta). Finally, we perform an experimental comparison of several enumeration algorithms for k <= 10 and k >= vertical bar V vertical bar - 3. (C) 2020 Elsevier B.V. All rights reserved.

A sub-constant error-probability low-degree test, and a sub-constant error-probability PCP characterization of NP

Completely derandomized self-adaptation in evolution strategies. This paper puts forward two useful methods for self-adaptation of the mutation distribution - the concepts of derandomization and cumulation. Principle shortcomings of the concept of mutative strategy parameter control and two levels of derandomization are reviewed. Basic demands on the self-adaptation of arbitrary (normal) mutation distributions are developed. Applying arbitrary, normal mutation distributions is equivalent to applying a general, linear problem encoding. The underlying objective of mutative strategy parameter control is roughly to favor previously selected mutation steps in the future. If this objective is pursued rigorously, a completely derandomized self-adaptation scheme results, which adapts arbitrary normal mutation distributions. This scheme, called covariance matrix adaptation (CMA), meets the previously stated demands. It can still be considerably improved by cumulation - utilizing an evolution path rather than single search steps. Simulations on various test functions reveal local and global search properties of the evolution strategy with and without covariance matrix adaptation. Their performances are comparable only on perfectly scaled functions. On badly scaled, non-separable functions usually a speed up factor of several orders of magnitude is observed. On moderately mis-scaled functions a speed up factor of three to ten can be expected.

A survey of socially interactive robots This paper reviews “socially interactive robots”: robots for which social human–robot interaction is important. We begin by discussing the context for socially interactive robots, emphasizing the relationship to other research fields and the different forms of “social robots”. We then present a taxonomy of design methods and system components used to build socially interactive robots. Finally, we describe the impact of these robots on humans and discuss open issues. An expanded version of this paper, which contains a survey and taxonomy of current applications, is available as a technical report [T. Fong, I. Nourbakhsh, K. Dautenhahn, A survey of socially interactive robots: concepts, design and applications, Technical Report No. CMU-RI-TR-02-29, Robotics Institute, Carnegie Mellon University, 2002].

Energy Efficiency Resource Allocation For D2d Communication Network Based On Relay Selection In order to solve the problem of spectrum resource shortage and energy consumption, we put forward a new model that combines with D2D communication and energy harvesting technology: energy harvesting-aided D2D communication network under the cognitive radio (EHA-CRD), where the D2D users harvest energy from the base station and the D2D source communicate with D2D destination by D2D relays. Our goals are to investigate the maximization energy efficiency (EE) of the network by joint time allocation and relay selection while taking into the constraints of the signal-to-noise ratio of D2D and the rates of the Cellular users. During this process, the energy collection time and communication time are randomly allocated. The maximization problem of EE can be divided into two sub-problems: (1) relay selection problem; (2) time optimization problem. For the first sub-problem, we propose a weighted sum maximum algorithm to select the best relay. For the last sub-problem, the EE maximization problem is non-convex problem with time. Thus, by using fractional programming theory, we transform it into a standard convex optimization problem, and we propose the optimization iterative algorithm to solve the convex optimization problem for obtaining the optimal solution. And, the simulation results show that the proposed relay selection algorithm and time optimization algorithm are significantly improved compared with the existing algorithms.

A communication robot in a shopping mall This paper reports our development of a communication robot for use in a shopping mall to provide shopping information, offer route guidance, and build rapport. In the development, the major difficulties included sensing human behaviors, conversation in a noisy daily environment, and the needs of unexpected miscellaneous knowledge in the conversation. We chose a networkrobot system approach, where a single robot's poor sensing capability and knowledge are supplemented by ubiquitous sensors and a human operator. The developed robot system detects a person with floor sensors to initiate interaction, identifies individuals with radio-frequency identification (RFID) tags, gives shopping information while chatting, and provides route guidance with deictic gestures. The robotwas partially teleoperated to avoid the difficulty of speech recognition as well as to furnish a new kind of knowledge that only humans can flexibly provide. The information supplied by a human operator was later used to increase the robot's autonomy. For 25 days in a shopping mall, we conducted a field trial and gathered 2642 interactions. A total of 235 participants signed up to use RFID tags and, later, provided questionnaire responses. The questionnaire results are promising in terms of the visitors' perceived acceptability as well as the encouragement of their shopping activities. The results of the teleoperation analysis revealed that the amount of teleoperation gradually decreased, which is also promising.

Minimum acceleration criterion with constraints implies bang-bang control as an underlying principle for optimal trajectories of arm reaching movements. Rapid arm-reaching movements serve as an excellent test bed for any theory about trajectory formation. How are these movements planned? A minimum acceleration criterion has been examined in the past, and the solution obtained, based on the Euler-Poisson equation, failed to predict that the hand would begin and end the movement at rest (i.e., with zero acceleration). Therefore, this criterion was rejected in favor of the minimum jerk, which was proved to be successful in describing many features of human movements. This letter follows an alternative approach and solves the minimum acceleration problem with constraints using Pontryagin's minimum principle. We use the minimum principle to obtain minimum acceleration trajectories and use the jerk as a control signal. In order to find a solution that does not include nonphysiological impulse functions, constraints on the maximum and minimum jerk values are assumed. The analytical solution provides a three-phase piecewise constant jerk signal (bang-bang control) where the magnitude of the jerk and the two switching times depend on the magnitude of the maximum and minimum available jerk values. This result fits the observed trajectories of reaching movements and takes into account both the extrinsic coordinates and the muscle limitations in a single framework. The minimum acceleration with constraints principle is discussed as a unifying approach for many observations about the neural control of movements.

An Automatic Screening Approach for Obstructive Sleep Apnea Diagnosis Based on Single-Lead Electrocardiogram Traditional approaches for obstructive sleep apnea (OSA) diagnosis are apt to using multiple channels of physiological signals to detect apnea events by dividing the signals into equal-length segments, which may lead to incorrect apnea event detection and weaken the performance of OSA diagnosis. This paper proposes an automatic-segmentation-based screening approach with the single channel of Electrocardiogram (ECG) signal for OSA subject diagnosis, and the main work of the proposed approach lies in three aspects: (i) an automatic signal segmentation algorithm is adopted for signal segmentation instead of the equal-length segmentation rule; (ii) a local median filter is improved for reduction of the unexpected RR intervals before signal segmentation; (iii) the designed OSA severity index and additional admission information of OSA suspects are plugged into support vector machine (SVM) for OSA subject diagnosis. A real clinical example from PhysioNet database is provided to validate the proposed approach and an average accuracy of 97.41% for subject diagnosis is obtained which demonstrates the effectiveness for OSA diagnosis.

A robust medical image watermarking against salt and pepper noise for brain MRI images. The ever-growing numbers of medical digital images and the need to share them among specialists and hospitals for better and more accurate diagnosis require that patients' privacy be protected. During the transmission of medical images between hospitals or specialists through the network, the main priority is to protect a patient's documents against any act of tampering by unauthorised individuals. Because of this, there is a need for medical image authentication scheme to enable proper diagnosis on patient. In addition, medical images are also susceptible to salt and pepper impulse noise through the transmission in communication channels. This noise may also be intentionally used by the invaders to corrupt the embedded watermarks inside the medical images. A common drawback of existing watermarking methods is their weakness against salt and pepper noise. The research carried out in this work addresses the issue of designing a new watermarking method that can withstand high density of salt and pepper noise for brain MRI images. For this purpose, combination of a spatial domain watermarking method, channel coding and noise filtering schemes are used. The region of non-interest (RONI) of MRI images from five different databases are used as embedding area and electronic patient record (EPR) is considered as embedded data. The quality of watermarked image is evaluated using Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM), and the accuracy of the extracted watermark is assessed in terms of Bit Error Rate (BER).

Convert Harm Into Benefit: A Coordination-Learning Based Dynamic Spectrum Anti-Jamming Approach This paper mainly investigates the multi-user anti-jamming spectrum access problem. Using the idea of “converting harm into benefit,” the malicious jamming signals projected by the enemy are utilized by the users as the coordination signals to guide spectrum coordination. An “internal coordination-external confrontation” multi-user anti-jamming access game model is constructed, and the existence of Nash equilibrium (NE) as well as correlated equilibrium (CE) is demonstrated. A coordination-learning based anti-jamming spectrum access algorithm (CLASA) is designed to achieve the CE of the game. Simulation results show the convergence, and effectiveness of the proposed CLASA algorithm, and indicate that our approach can help users confront the malicious jammer, and coordinate internal spectrum access simultaneously without information exchange. Last but not least, the fairness of the proposed approach under different jamming attack patterns is analyzed, which illustrates that this approach provides fair anti-jamming spectrum access opportunities under complicated jamming pattern.

Deep Learning in Mobile and Wireless Networking: A Survey. The rapid uptake of mobile devices and the rising popularity of mobile applications and services pose unprecedented demands on mobile and wireless networking infrastructure. Upcoming 5G systems are evolving to support exploding mobile traffic volumes, real-time extraction of fine-grained analytics, and agile management of network resources, so as to maximize user experience. Fulfilling these tasks is challenging, as mobile environments are increasingly complex, heterogeneous, and evolving. One potential solution is to resort to advanced machine learning techniques, in order to help manage the rise in data volumes and algorithm-driven applications. The recent success of deep learning underpins new and powerful tools that tackle problems in this space. In this paper we bridge the gap between deep learning and mobile and wireless networking research, by presenting a comprehensive survey of the crossovers between the two areas. We first briefly introduce essential background and state-of-the-art in deep learning techniques with potential applications to networking. We then discuss several techniques and platforms that facilitate the efficient deployment of deep learning onto mobile systems. Subsequently, we provide an encyclopedic review of mobile and wireless networking research based on deep learning, which we categorize by different domains. Drawing from our experience, we discuss how to tailor deep learning to mobile environments. We complete this survey by pinpointing current challenges and open future directions for research.

BeCome: Blockchain-Enabled Computation Offloading for IoT in Mobile Edge Computing Benefiting from the real-time processing ability of edge computing, computing tasks requested by smart devices in the Internet of Things are offloaded to edge computing devices (ECDs) for implementation. However, ECDs are often overloaded or underloaded with disproportionate resource requests. In addition, during the process of task offloading, the transmitted information is vulnerable, which can result in data incompleteness. In view of this challenge, a blockchain-enabled computation offloading method, named BeCome, is proposed in this article. Blockchain technology is employed in edge computing to ensure data integrity. Then, the nondominated sorting genetic algorithm III is adopted to generate strategies for balanced resource allocation. Furthermore, simple additive weighting and multicriteria decision making are utilized to identify the optimal offloading strategy. Finally, performance evaluations of BeCome are given through simulation experiments.

Mobile cloud computing [Guest Edotorial] Mobile Cloud Computing refers to an infrastructure where both the data storage and the data processing occur outside of the mobile device. Mobile cloud applications move the computing power and data storage away from mobile devices and into the cloud, bringing applications and mobile computing not only to smartphone users but also to a much broader range of mobile subscribers.

Multiuser Joint Task Offloading and Resource Optimization in Proximate Clouds Proximate cloud computing enables computationally intensive applications on mobile devices, providing a rich user experience. However, remote resource bottlenecks limit the scalability of offloading, requiring optimization of the offloading decision and resource utilization. To this end, in this paper, we leverage the variability in capabilities of mobile devices and user preferences. Our system u...

Deep Reinforcement Learning for Collaborative Edge Computing in Vehicular Networks Mobile edge computing (MEC) is a promising technology to support mission-critical vehicular applications, such as intelligent path planning and safety applications. In this paper, a collaborative edge computing framework is developed to reduce the computing service latency and improve service reliability for vehicular networks. First, a task partition and scheduling algorithm (TPSA) is proposed to decide the workload allocation and schedule the execution order of the tasks offloaded to the edge servers given a computation offloading strategy. Second, an artificial intelligence (AI) based collaborative computing approach is developed to determine the task offloading, computing, and result delivery policy for vehicles. Specifically, the offloading and computing problem is formulated as a Markov decision process. A deep reinforcement learning technique, i.e., deep deterministic policy gradient, is adopted to find the optimal solution in a complex urban transportation network. By our approach, the service cost, which includes computing service latency and service failure penalty, can be minimized via the optimal workload assignment and server selection in collaborative computing. Simulation results show that the proposed AI-based collaborative computing approach can adapt to a highly dynamic environment with outstanding performance.

Deep Reinforcement Learning for Mobile Edge Caching: Review, New Features, and Open Issues. Mobile edge caching is a promising technique to reduce network traffic and improve the quality of experience of mobile users. However, mobile edge caching is a challenging decision making problem with unknown future content popularity and complex network characteristics. In this article, we advocate the use of DRL to solve mobile edge caching problems by presenting an overview of recent works on m...

An Energy-Efficient Mobile-Sink Path Selection Strategy for Wireless Sensor Networks Several studies have demonstrated the benefits of using a mobile sink to reduce the energy consumption of nodes and to prevent the formation of energy holes in wireless sensor networks (WSNs). However, these benefits are dependent on the path taken by the mobile sink, particularly in delay-sensitive applications, as all sensed data must be collected within a given time constraint. An approach proposed to address this challenge is to form a hybrid moving pattern in which a mobile-sink node only visits rendezvous points (RPs), as opposed to all nodes. Sensor nodes that are not RPs forward their sensed data via multihopping to the nearest RP. The fundamental problem then becomes computing a tour that visits all these RPs within a given delay bound. Identifying the optimal tour, however, is an NP-hard problem. To address this problem, a heuristic called weighted rendezvous planning (WRP) is proposed, whereby each sensor node is assigned a weight corresponding to its hop distance from the tour and the number of data packets that it forwards to the closest RP. WRP is validated via extensive computer simulation, and our results demonstrate that WRP enables a mobile sink to retrieve all sensed data within a given deadline while conserving the energy expenditure of sensor nodes. More specifically, WRP reduces energy consumption by 22% and increases network lifetime by 44%, as compared with existing algorithms.

Massive MIMO for next generation wireless systems Multi-user MIMO offers big advantages over conventional point-to-point MIMO: it works with cheap single-antenna terminals, a rich scattering environment is not required, and resource allocation is simplified because every active terminal utilizes all of the time-frequency bins. However, multi-user MIMO, as originally envisioned, with roughly equal numbers of service antennas and terminals and frequency-division duplex operation, is not a scalable technology. Massive MIMO (also known as large-scale antenna systems, very large MIMO, hyper MIMO, full-dimension MIMO, and ARGOS) makes a clean break with current practice through the use of a large excess of service antennas over active terminals and time-division duplex operation. Extra antennas help by focusing energy into ever smaller regions of space to bring huge improvements in throughput and radiated energy efficiency. Other benefits of massive MIMO include extensive use of inexpensive low-power components, reduced latency, simplification of the MAC layer, and robustness against intentional jamming. The anticipated throughput depends on the propagation environment providing asymptotically orthogonal channels to the terminals, but so far experiments have not disclosed any limitations in this regard. While massive MIMO renders many traditional research problems irrelevant, it uncovers entirely new problems that urgently need attention: the challenge of making many low-cost low-precision components that work effectively together, acquisition and synchronization for newly joined terminals, the exploitation of extra degrees of freedom provided by the excess of service antennas, reducing internal power consumption to achieve total energy efficiency reductions, and finding new deployment scenarios. This article presents an overview of the massive MIMO concept and contemporary research on the topic.

Task Offloading in Vehicular Edge Computing Networks: A Load-Balancing Solution Recently, the rapid advance of vehicular networks has led to the emergence of diverse delay-sensitive vehicular applications such as automatic driving, auto navigation. Note that existing resource-constrained vehicles cannot adequately meet these demands on low / ultra-low latency. By offloading parts of the vehicles’ compute-intensive tasks to the edge servers in proximity, mobile edge computing is envisioned as a promising paradigm, giving rise to the vehicular edge computing networks (VECNs). However, most existing works on task offloading in VECNs did not take the load balancing of the computation resources at the edge servers into account. To address these issues and given the high dynamics of vehicular networks, we introduce fiber-wireless (FiWi) technology to enhance VECNs, due to its advantages on centralized network management and supporting multiple communication techniques. Aiming to minimize the processing delay of the vehicles’ computation tasks, we propose a software-defined networking (SDN) based load-balancing task offloading scheme in FiWi enhanced VECNs, where SDN is introduced to provide supports for the centralized network and vehicle information management. Extensive analysis and numerical results corroborate that our proposed load-balancing scheme can achieve superior performance on processing delay reduction by utilizing the edge servers’ computation resources more efficiently.

Supervisory control of fuzzy discrete event systems: a formal approach. Fuzzy discrete event systems (DESs) were proposed recently by Lin and Ying [19], which may better cope with the real-world problems of fuzziness, impreciseness, and subjectivity such as those in biomedicine. As a continuation of [19], in this paper, we further develop fuzzy DESs by dealing with supervisory control of fuzzy DESs. More specifically: 1) we reformulate the parallel composition of crisp DESs, and then define the parallel composition of fuzzy DESs that is equivalent to that in [19]. Max-product and max-min automata for modeling fuzzy DESs are considered, 2) we deal with a number of fundamental problems regarding supervisory control of fuzzy DESs, particularly demonstrate controllability theorem and nonblocking controllability theorem of fuzzy DESs, and thus, present the conditions for the existence of supervisors in fuzzy DESs; 3) we analyze the complexity for presenting a uniform criterion to test the fuzzy controllability condition of fuzzy DESs modeled by max-product automata; in particular, we present in detail a general computing method for checking whether or not the fuzzy controllability condition holds, if max-min automata are used to model fuzzy DESs, and by means of this method we can search for all possible fuzzy states reachable from initial fuzzy state in max-min automata. Also, we introduce the fuzzy n-controllability condition for some practical problems, and 4) a number of examples serving to illustrate the applications of the derived results and methods are described; some basic properties related to supervisory control of fuzzy DESs are investigated. To conclude, some related issues are raised for further consideration.

A Model for Understanding How Virtual Reality Aids Complex Conceptual Learning Designers and evaluators of immersive virtual reality systems have many ideas concerning how virtual reality can facilitate learning. However, we have little information concerning which of virtual reality's features provide the most leverage for enhancing understanding or how to customize those affordances for different learning environments. In part, this reflects the truly complex nature of learning. Features of a learning environment do not act in isolation; other factors such as the concepts or skills to be learned, individual characteristics, the learning experience, and the interaction experience all play a role in shaping the learning process and its outcomes. Through Project Science Space, we have been trying to identify, use, and evaluate immersive virtual reality's affordances as a means to facilitate the mastery of complex, abstract concepts. In doing so, we are beginning to understand the interplay between virtual reality's features and other important factors in shaping the learning process and learning outcomes for this type of material. In this paper, we present a general model that describes how we think these factors work together and discuss some of the lessons we are learning about virtual reality's affordances in the context of this model for complex conceptual learning.

Dynamic Management of Virtual Infrastructures Cloud infrastructures are becoming an appropriate solution to address the computational needs of scientific applications. However, the use of public or on-premises Infrastructure as a Service (IaaS) clouds requires users to have non-trivial system administration skills. Resource provisioning systems provide facilities to choose the most suitable Virtual Machine Images (VMI) and basic configuration of multiple instances and subnetworks. Other tasks such as the configuration of cluster services, computational frameworks or specific applications are not trivial on the cloud, and normally users have to manually select the VMI that best fits, including undesired additional services and software packages. This paper presents a set of components that ease the access and the usability of IaaS clouds by automating the VMI selection, deployment, configuration, software installation, monitoring and update of Virtual Appliances. It supports APIs from a large number of virtual platforms, making user applications cloud-agnostic. In addition it integrates a contextualization system to enable the installation and configuration of all the user required applications providing the user with a fully functional infrastructure. Therefore, golden VMIs and configuration recipes can be easily reused across different deployments. Moreover, the contextualization agent included in the framework supports horizontal (increase/decrease the number of resources) and vertical (increase/decrease resources within a running Virtual Machine) by properly reconfiguring the software installed, considering the configuration of the multiple resources running. This paves the way for automatic virtual infrastructure deployment, customization and elastic modification at runtime for IaaS clouds.

Adversarial Example Generation with Syntactically Controlled Paraphrase Networks. We propose syntactically controlled paraphrase networks (SCPNs) and use them to generate adversarial examples. Given a sentence and a target syntactic form (e.g., a constituency parse), SCPNs are trained to produce a paraphrase of the sentence with the desired syntax. We show it is possible to create training data for this task by first doing backtranslation at a very large scale, and then using a parser to label the syntactic transformations that naturally occur during this process. Such data allows us to train a neural encoder-decoder model with extra inputs to specify the target syntax. A combination of automated and human evaluations show that SCPNs generate paraphrases that follow their target specifications without decreasing paraphrase quality when compared to baseline (uncontrolled) paraphrase systems. Furthermore, they are more capable of generating syntactically adversarial examples that both (1) fool pretrained models and (2) improve the robustness of these models to syntactic variation when used to augment their training data.

Ethical Considerations Of Applying Robots In Kindergarten Settings: Towards An Approach From A Macroperspective In child-robot interaction (cHRI) research, many studies pursue the goal to develop interactive systems that can be applied in everyday settings. For early education, increasingly, the setting of a kindergarten is targeted. However, when cHRI and research are brought into a kindergarten, a range of ethical and related procedural aspects have to be considered and dealt with. While ethical models elaborated within other human-robot interaction settings, e.g., assisted living contexts, can provide some important indicators for relevant issues, we argue that it is important to start developing a systematic approach to identify and tackle those ethical issues which rise with cHRI in kindergarten settings on a more global level and address the impact of the technology from a macroperspective beyond the effects on the individual. Based on our experience in conducting studies with children in general and pedagogical considerations on the role of the institution of kindergarten in specific, in this paper, we enfold some relevant aspects that have barely been addressed in an explicit way in current cHRI research. Four areas are analyzed and key ethical issues are identified in each area: (1) the institutional setting of a kindergarten, (2) children as a vulnerable group, (3) the caregivers' role, and (4) pedagogical concepts. With our considerations, we aim at (i) broadening the methodology of the current studies within the area of cHRI, (ii) revalidate it based on our comprehensive empirical experience with research in kindergarten settings, both laboratory and real-world contexts, and (iii) provide a framework for the development of a more systematic approach to address the ethical issues in cHRI research within kindergarten settings.

Precomputing Oblivious Transfer Alice and Bob are too untrusting of computer scientists to let their privacy depend on unproven assumptions such as the existence of one-way functions. Firm believers in Schrödinger and Heisenberg, they might accept a quantum OT device, but IBM’s prototype is not yet portable. Instead, as part of their prenuptial agreement, they decide to visit IBM and perform some OT’s in advance, so that any later divorces, coin-flipping or other important interactions can be done more conveniently, without needing expensive third parties. Unfortunately, OT can’t be done in advance in a direct way, because even though Bob might not know what bit Alice will later send (even if she first sends a random bit and later corrects it, for example), he would already know which bit or bits he will receive. We address the problem of precomputing oblivious transfer and show that OT can be precomputed at a cost of Θ(κ) prior transfers (a tight bound). In contrast, we show that variants of OT, such as one-out-of-two OT, can be precomputed using only one prior transfer. Finally, we show that all variants can be reduced to a single precomputed one-out-of-two oblivious transfer.

Priced Oblivious Transfer: How to Sell Digital Goods We consider the question of protecting the privacy of customers buying digital goods. More specifically, our goal is to allow a buyer to purchase digital goods from a vendor without letting the vendor learn what, and to the extent possible also when and how much, it is buying. We propose solutions which allow the buyer, after making an initial deposit, to engage in an unlimited number of priced oblivious-transfer protocols, satisfying the following requirements: As long as the buyer's balance contains sufficient funds, it will successfully retrieve the selected item and its balance will be debited by the item's price. However, the buyer should be unable to retrieve an item whose cost exceeds its remaining balance. The vendor should learn nothing except what must inevitably be learned, namely, the amount of interaction and the initial deposit amount (which imply upper bounds on the quantity and total price of all information obtained by the buyer). In particular, the vendor should be unable to learn what the buyer's current balance is or when it actually runs out of its funds. The technical tools we develop, in the process of solving this problem, seem to be of independent interest. In particular, we present the first one-round (two-pass) protocol for oblivious transfer that does not rely on the random oracle model (a very similar protocol was independently proposed by Naor and Pinkas [21]). This protocol is a special case of a more general "conditional disclosure" methodology, which extends a previous approach from [11] and adapts it to the 2-party setting.

Efficient k-out-of-n oblivious transfer schemes with adaptive and non-adaptive queries In this paper we propose efficient two-round k-out-of-n oblivious transfer schemes, in which R sends O(k) messages to S, and S sends O(n) messages back to R. The computation cost of R and S is reasonable. The choices of R are unconditionally secure. For the basic scheme, the secrecy of unchosen messages is guaranteed if the Decisional Diffie-Hellman problem is hard. When k=1, our basic scheme is as efficient as the most efficient 1-out-of-n oblivious transfer scheme. Our schemes have the nice property of universal parameters, that is each pair of R and S need neither hold any secret key nor perform any prior setup (initialization). The system parameters can be used by all senders and receivers without any trapdoor specification. Our k-out-of-n oblivious transfer schemes are the most efficient ones in terms of the communication cost, in both rounds and the number of messages. Moreover, one of our schemes can be extended in a straightforward way to an adaptivek-out-of-n oblivious transfer scheme, which allows the receiver R to choose the messages one by one adaptively. In our adaptive-query scheme, S sends O(n) messages to R in one round in the commitment phase. For each query of R, only O(1) messages are exchanged and O(1) operations are performed. In fact, the number k of queries need not be pre-fixed or known beforehand. This makes our scheme highly flexible.

MeDShare: Trust-Less Medical Data Sharing Among Cloud Service Providers via Blockchain. The dissemination of patients' medical records results in diverse risks to patients' privacy as malicious activities on these records cause severe damage to the reputation, finances, and so on of all parties related directly or indirectly to the data. Current methods to effectively manage and protect medical records have been proved to be insufficient. In this paper, we propose MeDShare, a system that addresses the issue of medical data sharing among medical big data custodians in a trust-less environment. The system is blockchain-based and provides data provenance, auditing, and control for shared medical data in cloud repositories among big data entities. MeDShare monitors entities that access data for malicious use from a data custodian system. In MeDShare, data transitions and sharing from one entity to the other, along with all actions performed on the MeDShare system, are recorded in a tamper-proof manner. The design employs smart contracts and an access control mechanism to effectively track the behavior of the data and revoke access to offending entities on detection of violation of permissions on data. The performance of MeDShare is comparable to current cutting edge solutions to data sharing among cloud service providers. By implementing MeDShare, cloud service providers and other data guardians will be able to achieve data provenance and auditing while sharing medical data with entities such as research and medical institutions with minimal risk to data privacy.

Personal Data Trading Scheme for Data Brokers in IoT Data Marketplaces. With the widespread use of the Internet of Things, data-driven services take the lead of both online and off-line businesses. Especially, personal data draw heavy attention of service providers because of the usefulness in value-added services. With the emerging big-data technology, a data broker appears, which exploits and sells personal data about individuals to other third parties. Due to little transparency between providers and brokers/consumers, people think that the current ecosystem is not trustworthy, and new regulations with strengthening the rights of individuals were introduced. Therefore, people have an interest in their privacy valuation. In this sense, the willingness-to-sell (WTS) of providers becomes one of the important aspects for data brokers; however, conventional studies have mainly focused on the willingnessto-buy (WTB) of consumers. Therefore, this paper proposes an optimized trading model for data brokers who buy personal data with proper incentives based on the WTS, and they sell valuable information from the refined dataset by considering the WTB and the dataset quality. This paper shows that the proposed model has a global optimal point by the convex optimization technique and proposes a gradient ascent-based algorithm. Consequently, it shows that the proposed model is feasible even if the data brokers spend costs to gather personal data.

How to Use Bitcoin to Design Fair Protocols. We study a model of fairness in secure computation in which an adversarial party that aborts on receiving output is forced to pay a mutually predefined monetary penalty. We then show how the Bitcoin network can be used to achieve the above notion of fairness in the twoparty as well as the multiparty setting (with a dishonest majority). In particular, we propose new ideal functionalities and protocols for fair secure computation and fair lottery in this model. One of our main contributions is the definition of an ideal primitive, which we call F-CR(star) (CR stands for "claim-or-refund"), that formalizes and abstracts the exact properties we require from the Bitcoin network to achieve our goals. Naturally, this abstraction allows us to design fair protocols in a hybrid model in which parties have access to the F-CR(star) functionality, and is otherwise independent of the Bitcoin ecosystem. We also show an efficient realization of F-CR(star) that requires only two Bitcoin transactions to be made on the network. Our constructions also enjoy high efficiency. In a multiparty setting, our protocols only require a constant number of calls to F-CR(star) per party on top of a standard multiparty secure computation protocol. Our fair multiparty lottery protocol improves over previous solutions which required a quadratic number of Bitcoin transactions.

Dynamic Fully Homomorphic encryption-based Merkle Tree for lightweight streaming authenticated data structures. Fully Homomorphic encryption-based Merkle Tree (FHMT) is a novel technique for streaming authenticated data structures (SADS) to achieve the streaming verifiable computation. By leveraging the computing capability of fully homomorphic encryption, FHMT shifts almost all of the computation tasks to the server, reaching nearly no overhead for the client. Therefore, FHMT is an important technique to construct a more efficient lightweight ADS for resource-limited clients. But the typical FHMT cannot support the dynamic scenario very well because it cannot expend freely since its height is fixed. We now present our fully dynamic FHMT construction, which is a construction that is able to authenticate an unbounded number of data elements and improves upon the state-of-the-art in terms of computational overhead. We divided the algorithms of the DFHMT with the following phases: initialization, insertion, tree expansion, query and verification. The DFHMT removes the drawbacks of the static FHMT. In the initialization phase, it is not required for the scale of the tree to be determined, and the scale of the tree can be adaptively expanded during the data-appending phase. This feature is more suitable for streaming data environments. We analyzed the security of the DFHMT, and point out that DFHMT has the same security with FHMT. The storage, communication and computation overhead of DFHMT is also analyzed, the results show that the client uses simple numerical multiplications and additions to replace hash operations, which reduces the computational burden of the client; the length of the authentication path in DFHMT is shorter than FHMT, which reduces storage and communication overhead. The performance of DFHMT was compared with other construction techniques of SADS via some tests, the results show that DFHMT strikes the performance balance between the client and server, which has some performance advantage for lightweight devices.

A standalone RFID Indoor Positioning System Using Passive Tags Indoor positioning systems (IPSs) locate objects in closed structures such as office buildings, hospitals, stores, factories, and warehouses, where Global Positioning System devices generally do not work. Most available systems apply wireless concepts, optical tracking, and/or ultrasound. This paper presents a standalone IPS using radio frequency identification (RFID) technology. The concept is ba...

A Private and Efficient Mechanism for Data Uploading in Smart Cyber-Physical Systems. To provide fine-grained access to different dimensions of the physical world, the data uploading in smart cyber-physical systems suffers novel challenges on both energy conservation and privacy preservation. It is always critical for participants to consume as little energy as possible for data uploading. However, simply pursuing energy efficiency may lead to extreme disclosure of private informat...

Image forgery detection We are undoubtedly living in an age where we are exposed to a remarkable array of visual imagery. While we may have historically had confidence in the integrity of this imagery, today&#39;s digital technology has begun to erode this trust. From the tabloid magazines to the fashion industry and in mainstream media outlets, scientific journals, political campaigns, courtrooms, and the photo hoaxes that ...

IntrospectiveViews: an interface for scrutinizing semantic user models User models are a key component for user-adaptive systems They represent information about users such as interests, expertise, goals, traits, etc This information is used to achieve various adaptation effects, e.g., recommending relevant documents or products To ensure acceptance by users, these models need to be scrutable, i.e., users must be able to view and alter them to understand and if necessary correct the assumptions the system makes about the user However, in most existing systems, this goal is not met In this paper, we introduce IntrospectiveViews, an interface that enables the user to view and edit her user model Furthermore, we present the results of a formative evaluation that show the importance users give in general to different aspects of scrutable user models and also substantiate our claim that IntrospectiveViews is an appropriate realization of an interface to such models.

Placing Virtual Machines to Optimize Cloud Gaming Experience Optimizing cloud gaming experience is no easy task due to the complex tradeoff between gamer quality of experience (QoE) and provider net profit. We tackle the challenge and study an optimization problem to maximize the cloud gaming provider's total profit while achieving just-good-enough QoE. We conduct measurement studies to derive the QoE and performance models. We formulate and optimally solve the problem. The optimization problem has exponential running time, and we develop an efficient heuristic algorithm. We also present an alternative formulation and algorithms for closed cloud gaming services with dedicated infrastructures, where the profit is not a concern and overall gaming QoE needs to be maximized. We present a prototype system and testbed using off-the-shelf virtualization software, to demonstrate the practicality and efficiency of our algorithms. Our experience on realizing the testbed sheds some lights on how cloud gaming providers may build up their own profitable services. Last, we conduct extensive trace-driven simulations to evaluate our proposed algorithms. The simulation results show that the proposed heuristic algorithms: (i) produce close-to-optimal solutions, (ii) scale to large cloud gaming services with 20,000 servers and 40,000 gamers, and (iii) outperform the state-of-the-art placement heuristic, e.g., by up to 3.5 times in terms of net profits.

A multi-objective and PSO based energy efficient path design for mobile sink in wireless sensor networks. Data collection through mobile sink (MS) in wireless sensor networks (WSNs) is an effective solution to the hot-spot or sink-hole problem caused by multi-hop routing using the static sink. Rendezvous point (RP) based MS path design is a common and popular technique used in this regard. However, design of the optimal path is a well-known NP-hard problem. Therefore, an evolutionary approach like multi-objective particle swarm optimization (MOPSO) can prove to be a very promising and reasonable approach to solve the same. In this paper, we first present a Linear Programming formulation for the stated problem and then, propose an MOPSO-based algorithm to design an energy efficient trajectory for the MS. The algorithm is presented with an efficient particle encoding scheme and derivation of a proficient multi-objective fitness function. We use Pareto dominance in MOPSO for obtaining both local and global best guides for each particle. We carry out rigorous simulation experiments on the proposed algorithm and compare the results with two existing algorithms namely, tree cluster based data gathering algorithm (TCBDGA) and energy aware sink relocation (EASR). The results demonstrate that the proposed algorithm performs better than both of them in terms of various performance metrics. The results are also validated through the statistical test, analysis of variance (ANOVA) and its least significant difference (LSD) post hoc analysis.

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

Learning from Multiple Cities: A Meta-Learning Approach for Spatial-Temporal Prediction Spatial-temporal prediction is a fundamental problem for constructing smart city, which is useful for tasks such as traffic control, taxi dispatching, and environment policy making. Due to data collection mechanism, it is common to see data collection with unbalanced spatial distributions. For example, some cities may release taxi data for multiple years while others only release a few days of data; some regions may have constant water quality data monitored by sensors whereas some regions only have a small collection of water samples. In this paper, we tackle the problem of spatial-temporal prediction for the cities with only a short period of data collection. We aim to utilize the long-period data from other cities via transfer learning. Different from previous studies that transfer knowledge from one single source city to a target city, we are the first to leverage information from multiple cities to increase the stability of transfer. Specifically, our proposed model is designed as a spatial-temporal network with a meta-learning paradigm. The meta-learning paradigm learns a well-generalized initialization of the spatial-temporal network, which can be effectively adapted to target cities. In addition, a pattern-based spatial-temporal memory is designed to distill long-term temporal information (i.e., periodicity). We conduct extensive experiments on two tasks: traffic (taxi and bike) prediction and water quality prediction. The experiments demonstrate the effectiveness of our proposed model over several competitive baseline models.

Forecasting holiday daily tourist flow based on seasonal support vector regression with adaptive genetic algorithm. •The model of support vector regression with adaptive genetic algorithm and the seasonal mechanism is proposed.•Parameters selection and seasonal adjustment should be carefully selected.•We focus on latest and representative holiday daily data in China.•Two experiments are used to prove the effect of the model.•The AGASSVR is superior to AGA-SVR and BPNN.

Regression conformal prediction with random forests Regression conformal prediction produces prediction intervals that are valid, i.e., the probability of excluding the correct target value is bounded by a predefined confidence level. The most important criterion when comparing conformal regressors is efficiency; the prediction intervals should be as tight (informative) as possible. In this study, the use of random forests as the underlying model for regression conformal prediction is investigated and compared to existing state-of-the-art techniques, which are based on neural networks and k-nearest neighbors. In addition to their robust predictive performance, random forests allow for determining the size of the prediction intervals by using out-of-bag estimates instead of requiring a separate calibration set. An extensive empirical investigation, using 33 publicly available data sets, was undertaken to compare the use of random forests to existing state-of-the-art conformal predictors. The results show that the suggested approach, on almost all confidence levels and using both standard and normalized nonconformity functions, produced significantly more efficient conformal predictors than the existing alternatives.

Learning to Predict Bus Arrival Time From Heterogeneous Measurements via Recurrent Neural Network Bus arrival time prediction intends to improve the level of the services provided by transportation agencies. Intuitively, many stochastic factors affect the predictability of the arrival time, e.g., weather and local events. Moreover, the arrival time prediction for a current station is closely correlated with that of multiple passed stations. Motivated by the observations above, this paper propo...

Hybrid Spatio-Temporal Graph Convolutional Network: Improving Traffic Prediction with Navigation Data Traffic forecasting has recently attracted increasing interest due to the popularity of online navigation services, ridesharing and smart city projects. Owing to the non-stationary nature of road traffic, forecasting accuracy is fundamentally limited by the lack of contextual information. To address this issue, we propose the Hybrid Spatio-Temporal Graph Convolutional Network (H-STGCN), which is able to "deduce" future travel time by exploiting the data of upcoming traffic volume. Specifically, we propose an algorithm to acquire the upcoming traffic volume from an online navigation engine. Taking advantage of the piecewise-linear flow-density relationship, a novel transformer structure converts the upcoming volume into its equivalent in travel time. We combine this signal with the commonly-utilized travel-time signal, and then apply graph convolution to capture the spatial dependency. Particularly, we construct a compound adjacency matrix which reflects the innate traffic proximity. We conduct extensive experiments on real-world datasets. The results show that H-STGCN remarkably outperforms state-of-the-art methods in various metrics, especially for the prediction of non-recurring congestion.

Long-Term Traffic Speed Prediction Based on Multiscale Spatio-Temporal Feature Learning Network Speed plays a significant role in evaluating the evolution of traffic status, and predicting speed is one of the fundamental tasks for the intelligent transportation system. There exists a large number of works on speed forecast; however, the problem of long-term prediction for the next day is still not well addressed. In this paper, we propose a multiscale spatio-temporal feature learning network (MSTFLN) as the model to handle the challenging task of long-term traffic speed prediction for elevated highways. Raw traffic speed data collected from loop detectors every 5 min are transformed into spatial-temporal matrices; each matrix represents the one-day speed information, rows of the matrix indicate the numbers of loop detectors, and time intervals are denoted by columns. To predict the traffic speed of a certain day, nine speed matrices of three historical days with three different time scales are served as the input of MSTFLN. The proposed MSTFLN model consists of convolutional long short-term memories and convolutional neural networks. Experiments are evaluated using the data of three main elevated highways in Shanghai, China. The presented results demonstrate that our approach outperforms the state-of-the-art work and it can effectively predict the long-term speed information.

Deep Spatial-Temporal 3D Convolutional Neural Networks for Traffic Data Forecasting Reliable traffic prediction is critical to improve safety, stability, and efficiency of intelligent transportation systems. However, traffic prediction is a very challenging problem because traffic data are a typical type of spatio-temporal data, which simultaneously shows correlation and heterogeneity both in space and time. Most existing works can only capture the partial properties of traffic data and even assume that the effect of correlation on traffic prediction is globally invariable, resulting in inadequate modeling and unsatisfactory prediction performance. In this paper, we propose a novel end-to-end deep learning model, called ST-3DNet, for traffic raster data prediction. ST-3DNet introduces 3D convolutions to automatically capture the correlations of traffic data in both spatial and temporal dimensions. A novel recalibration (Rc) block is proposed to explicitly quantify the difference of the contributions of the correlations in space. Considering two kinds of temporal properties of traffic data, i.e., local patterns and long-term patterns, ST-3DNet employs two components consisting of 3D convolutions and Rc blocks to, respectively, model the two kinds of patterns and then aggregates them together in a weighted way for the final prediction. The experiments on several real-world traffic datasets, viz., traffic congestion data and crowd flows data, demonstrate that our ST-3DNet outperforms the state-of-the-art baselines.

Time-Aware Multivariate Nearest Neighbor Regression Methods for Traffic Flow Prediction Traffic flow prediction is a fundamental functionality of intelligent transportation systems. After presenting the state of the art, we focus on nearest neighbor regression methods, which are data-driven algorithms that are effective yet simple to implement. We try to strengthen their efficacy in two ways that are little explored in literature, i.e., by adopting a multivariate approach and by adding awareness of the time of the day. The combination of these two refinements, which represents a novelty, leads to the definition of a new class of methods that we call time-aware multivariate nearest neighbor regression (TaM-NNR) algorithms. To assess this class, we have used publicly available traffic data from a California highway. Computational results show the effectiveness of such algorithms in comparison with state-of-the-art parametric and non-parametric methods. In particular, they consistently perform better than their corresponding standard univariate versions. These facts highlight the importance of context elements in traffic prediction. The ideas presented here may be further investigated considering more context elements (e.g., weather conditions), more complex road topologies (e.g., urban networks), and different types of prediction methods.

On the security of public key protocols Recently the use of public key encryption to provide secure network communication has received considerable attention. Such public key systems are usually effective against passive eavesdroppers, who merely tap the lines and try to decipher the message. It has been pointed out, however, that an improperly designed protocol could be vulnerable to an active saboteur, one who may impersonate another user or alter the message being transmitted. Several models are formulated in which the security of protocols can be discussed precisely. Algorithms and characterizations that can be used to determine protocol security in these models are given.

Argos: practical many-antenna base stations Multi-user multiple-input multiple-output theory predicts manyfold capacity gains by leveraging many antennas on wireless base stations to serve multiple clients simultaneously through multi-user beamforming (MUBF). However, realizing a base station with a large number antennas is non-trivial, and has yet to be achieved in the real-world. We present the design, realization, and evaluation of Argos, the first reported base station architecture that is capable of serving many terminals simultaneously through MUBF with a large number of antennas (M >> 10). Designed for extreme flexibility and scalability, Argos exploits hierarchical and modular design principles, properly partitions baseband processing, and holistically considers real-time requirements of MUBF. Argos employs a novel, completely distributed, beamforming technique, as well as an internal calibration procedure to enable implicit beamforming with channel estimation cost independent of the number of base station antennas. We report an Argos prototype with 64 antennas and capable of serving 15 clients simultaneously. We experimentally demonstrate that by scaling from 1 to 64 antennas the prototype can achieve up to 6.7 fold capacity gains while using a mere 1/64th of the transmission power.

Load Scheduling and Dispatch for Aggregators of Plug-In Electric Vehicles This paper proposes an operating framework for aggregators of plug-in electric vehicles (PEVs). First, a minimum-cost load scheduling algorithm is designed, which determines the purchase of energy in the day-ahead market based on the forecast electricity price and PEV power demands. The same algorithm is applicable for negotiating bilateral contracts. Second, a dynamic dispatch algorithm is developed, used for distributing the purchased energy to PEVs on the operating day. Simulation results are used to evaluate the proposed algorithms, and to demonstrate the potential impact of an aggregated PEV fleet on the power system.

Stochastic Power Adaptation with Multiagent Reinforcement Learning for Cognitive Wireless Mesh Networks As the scarce spectrum resource is becoming overcrowded, cognitive radio indicates great flexibility to improve the spectrum efficiency by opportunistically accessing the authorized frequency bands. One of the critical challenges for operating such radios in a network is how to efficiently allocate transmission powers and frequency resource among the secondary users (SUs) while satisfying the quality-of-service constraints of the primary users. In this paper, we focus on the noncooperative power allocation problem in cognitive wireless mesh networks formed by a number of clusters with the consideration of energy efficiency. Due to the SUs' dynamic and spontaneous properties, the problem is modeled as a stochastic learning process. We first extend the single-agent $(Q)$-learning to a multiuser context, and then propose a conjecture-based multiagent $(Q)$-learning algorithm to achieve the optimal transmission strategies with only private and incomplete information. An intelligent SU performs $(Q)$-function updates based on the conjecture over the other SUs' stochastic behaviors. This learning algorithm provably converges given certain restrictions that arise during the learning procedure. Simulation experiments are used to verify the performance of our algorithm and demonstrate its effectiveness of improving the energy efficiency.

Substituting Motion Effects with Vibrotactile Effects for 4D Experiences. In this paper, we present two methods to substitute motion effects using vibrotactile effects in order to improve the 4D experiences of viewers. This work was motivated by the needs of more affordable 4D systems for individual users. Our sensory substitution algorithms convert motion commands to vibrotactile commands to a grid display that uses multiple actuators. While one method is based on the fundamental principle of vestibular feedback, the other method makes use of intuitive visually-based mapping from motion to vibrotactile stimulation. We carried out a user study and could confirm the effectiveness of our substitution methods in improving 4D experiences. To our knowledge, this is the first study that investigated the feasibility of replacing motion effects using much simpler and less expensive vibrotactile effects.

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

Self-Attentive Sequential Recommendation Sequential dynamics are a key feature of many modern recommender systems, which seek to capture the 'context' of users' activities on the basis of actions they have performed recently. To capture such patterns, two approaches have proliferated: Markov Chains (MCs) and Recurrent Neural Networks (RNNs). Markov Chains assume that a user's next action can be predicted on the basis of just their last (or last few) actions, while RNNs in principle allow for longer-term semantics to be uncovered. Generally speaking, MC-based methods perform best in extremely sparse datasets, where model parsimony is critical, while RNNs perform better in denser datasets where higher model complexity is affordable. The goal of our work is to balance these two goals, by proposing a self-attention based sequential model (SASRec) that allows us to capture long-term semantics (like an RNN), but, using an attention mechanism, makes its predictions based on relatively few actions (like an MC). At each time step, SASRec seeks to identify which items are 'relevant' from a user's action history, and use them to predict the next item. Extensive empirical studies show that our method outperforms various state-of-the-art sequential models (including MC/CNN/RNN-based approaches) on both sparse and dense datasets. Moreover, the model is an order of magnitude more efficient than comparable CNN/RNN-based models. Visualizations on attention weights also show how our model adaptively handles datasets with various density, and uncovers meaningful patterns in activity sequences.

Modeling Location-Based User Rating Profiles for Personalized Recommendation This article proposes LA-LDA, a location-aware probabilistic generative model that exploits location-based ratings to model user profiles and produce recommendations. Most of the existing recommendation models do not consider the spatial information of users or items; however, LA-LDA supports three classes of location-based ratings, namely spatial user ratings for nonspatial items, nonspatial user ratings for spatial items, and spatial user ratings for spatial items. LA-LDA consists of two components, ULA-LDA and ILA-LDA, which are designed to take into account user and item location information, respectively. The component ULA-LDA explicitly incorporates and quantifies the influence from local public preferences to produce recommendations by considering user home locations, whereas the component ILA-LDA recommends items that are closer in both taste and travel distance to the querying users by capturing item co-occurrence patterns, as well as item location co-occurrence patterns. The two components of LA-LDA can be applied either separately or collectively, depending on the available types of location-based ratings. To demonstrate the applicability and flexibility of the LA-LDA model, we deploy it to both top-k recommendation and cold start recommendation scenarios. Experimental evidence on large-scale real-world data, including the data from Gowalla (a location-based social network), DoubanEvent (an event-based social network), and MovieLens (a movie recommendation system), reveal that LA-LDA models user profiles more accurately by outperforming existing recommendation models for top-k recommendation and the cold start problem.

DeepMove: Predicting Human Mobility with Attentional Recurrent Networks. Human mobility prediction is of great importance for a wide spectrum of location-based applications. However, predicting mobility is not trivial because of three challenges: 1) the complex sequential transition regularities exhibited with time-dependent and high-order nature; 2) the multi-level periodicity of human mobility; and 3) the heterogeneity and sparsity of the collected trajectory data. In this paper, we propose DeepMove, an attentional recurrent network for mobility prediction from lengthy and sparse trajectories. In DeepMove, we first design a multi-modal embedding recurrent neural network to capture the complicated sequential transitions by jointly embedding the multiple factors that govern the human mobility. Then, we propose a historical attention model with two mechanisms to capture the multi-level periodicity in a principle way, which effectively utilizes the periodicity nature to augment the recurrent neural network for mobility prediction. We perform experiments on three representative real-life mobility datasets, and extensive evaluation results demonstrate that our model outperforms the state-of-the-art models by more than 10%. Moreover, compared with the state-of-the-art neural network models, DeepMove provides intuitive explanations into the prediction and sheds light on interpretable mobility prediction.

Deep Irregular Convolutional Residual LSTM for Urban Traffic Passenger Flows Prediction Urban traffic passenger flows prediction is practically important to facilitate many real applications including transportation management and public safety. Recently, deep learning based approaches are proposed to learn the spatio-temporal characteristics of the traffic passenger flows. However, it is still very challenging to handle some complex factors such as hybrid transportation lines, mixed traffic, transfer stations, and some extreme weathers. Considering the multi-channel and irregularity properties of urban traffic passenger flows in different transportation lines, a more efficient and fine-grained deep spatio-temporal feature learning model is necessary. In this paper, we propose a deep irregular convolutional residual LSTM network model called DST-ICRL for urban traffic passenger flows prediction. We first model the passenger flows among different traffic lines in a transportation network into multi-channel matrices analogous to the RGB pixel matrices of an image. Then, we propose a deep learning framework that integrates irregular convolutional residential network and LSTM units to learn the spatial-temporal feature representations. To fully utilize the historical passenger flows, we sample both the short-term and long-term historical traffic data, which can capture the periodicity and trend of the traffic passenger flows. In addition, we also fuse other external factors further to facilitate a real-time prediction. We conduct extensive experiments on different types of traffic passenger flows datasets including subway, taxi and bus flows in Beijing as well as bike flows in New York. The results show that the proposed DST-ICRL significantly outperforms both traditional and deep learning based urban traffic passenger flows prediction methods.

Dynamic Graph Convolutional Network For Long-Term Traffic Flow Prediction With Reinforcement Learning Exploiting deep learning techniques for traffic flow prediction has become increasingly widespread. Most existing studies combine CNN or GCN with recurrent neural network to extract the spatio-temporal features in traffic networks. The traffic networks can be naturally modeled as graphs which are effective to capture the topology and spatial correlations among road links. The issue is that the traffic network is dynamic due to the continuous changing of the traffic environment. Compared with the static graph, the dynamic graph can better reflect the spatio-temporal features of the traffic network. However, in practical applications, due to the limited accuracy and timeliness of data, it is hard to generate graph structures through frequent statistical data. Therefore, it is necessary to design a method to overcome data defects in traffic flow prediction. In this paper, we propose a long-term traffic flow prediction method based on dynamic graphs. The traffic network is modeled by dynamic traffic flow probability graphs, and graph convolution is performed on the dynamic graphs to learn spatial features, which are then combined with LSTM units to learn temporal features. In particular, we further propose to use graph convolutional policy network based on reinforcement learning to generate dynamic graphs when the dynamic graphs are incomplete due to the data sparsity i sue. By testing our method on city-bike data in New York City, it demonstrates that our model can achieve stable and effective long-term predictions of traffic flow, and can reduce the impact of data defects on prediction results. (c) 2021 Elsevier Inc. All rights reserved.

Map-matching for low-sampling-rate GPS trajectories Map-matching is the process of aligning a sequence of observed user positions with the road network on a digital map. It is a fundamental pre-processing step for many applications, such as moving object management, traffic flow analysis, and driving directions. In practice there exists huge amount of low-sampling-rate (e.g., one point every 2--5 minutes) GPS trajectories. Unfortunately, most current map-matching approaches only deal with high-sampling-rate (typically one point every 10--30s) GPS data, and become less effective for low-sampling-rate points as the uncertainty in data increases. In this paper, we propose a novel global map-matching algorithm called ST-Matching for low-sampling-rate GPS trajectories. ST-Matching considers (1) the spatial geometric and topological structures of the road network and (2) the temporal/speed constraints of the trajectories. Based on spatio-temporal analysis, a candidate graph is constructed from which the best matching path sequence is identified. We compare ST-Matching with the incremental algorithm and Average-Fréchet-Distance (AFD) based global map-matching algorithm. The experiments are performed both on synthetic and real dataset. The results show that our ST-matching algorithm significantly outperform incremental algorithm in terms of matching accuracy for low-sampling trajectories. Meanwhile, when compared with AFD-based global algorithm, ST-Matching also improves accuracy as well as running time.

Citywide traffic congestion estimation with social media. Conventional traffic congestion estimation approaches require the deployment of traffic sensors or large-scale probe vehicles. The high cost of deploying and maintaining these equipments largely limits their spatial-temporal coverage. This paper proposes an alternative solution with lower cost and wider spatial coverage by exploring traffic related information from Twitter. By regarding each Twitter user as a traffic monitoring sensor, various real-time traffic information can be collected freely from each corner of the city. However, there are two major challenges for this problem. Firstly, the congestion related information extracted directly from real-time tweets are very sparse due both to the low resolution of geographic location mentioned in the tweets and the inherent sparsity nature of Twitter data. Secondly, the traffic event information coming from Twitter can be multi-typed including congestion, accident, road construction, etc. It is non-trivial to model the potential impacts of diverse traffic events on traffic congestion. We propose to enrich the sparse real-time tweets from two directions: 1) mining the spatial and temporal correlations of the road segments in congestion from historical data, and 2) applying auxiliary information including social events and road features for help. We finally propose a coupled matrix and tensor factorization model to effectively integrate rich information for Citywide Traffic Congestion Eestimation (CTCE). Extensive evaluations on Twitter data and 500 million public passenger buses GPS data on nearly 700 mile roads of Chicago demonstrate the efficiency and effectiveness of the proposed approach.

Massive MIMO for next generation wireless systems Multi-user MIMO offers big advantages over conventional point-to-point MIMO: it works with cheap single-antenna terminals, a rich scattering environment is not required, and resource allocation is simplified because every active terminal utilizes all of the time-frequency bins. However, multi-user MIMO, as originally envisioned, with roughly equal numbers of service antennas and terminals and frequency-division duplex operation, is not a scalable technology. Massive MIMO (also known as large-scale antenna systems, very large MIMO, hyper MIMO, full-dimension MIMO, and ARGOS) makes a clean break with current practice through the use of a large excess of service antennas over active terminals and time-division duplex operation. Extra antennas help by focusing energy into ever smaller regions of space to bring huge improvements in throughput and radiated energy efficiency. Other benefits of massive MIMO include extensive use of inexpensive low-power components, reduced latency, simplification of the MAC layer, and robustness against intentional jamming. The anticipated throughput depends on the propagation environment providing asymptotically orthogonal channels to the terminals, but so far experiments have not disclosed any limitations in this regard. While massive MIMO renders many traditional research problems irrelevant, it uncovers entirely new problems that urgently need attention: the challenge of making many low-cost low-precision components that work effectively together, acquisition and synchronization for newly joined terminals, the exploitation of extra degrees of freedom provided by the excess of service antennas, reducing internal power consumption to achieve total energy efficiency reductions, and finding new deployment scenarios. This article presents an overview of the massive MIMO concept and contemporary research on the topic.

Energy-Efficient Resource Allocation for Mobile-Edge Computation Offloading. Mobile-edge computation offloading (MECO) off-loads intensive mobile computation to clouds located at the edges of cellular networks. Thereby, MECO is envisioned as a promising technique for prolonging the battery lives and enhancing the computation capacities of mobiles. In this paper, we study resource allocation for a multiuser MECO system based on time-division multiple access (TDMA) and orthogonal frequency-division multiple access (OFDMA). First, for the TDMA MECO system with infinite or finite cloud computation capacity, the optimal resource allocation is formulated as a convex optimization problem for minimizing the weighted sum mobile energy consumption under the constraint on computation latency. The optimal policy is proved to have a threshold-based structure with respect to a derived offloading priority function, which yields priorities for users according to their channel gains and local computing energy consumption. As a result, users with priorities above and below a given threshold perform complete and minimum offloading, respectively. Moreover, for the cloud with finite capacity, a sub-optimal resource-allocation algorithm is proposed to reduce the computation complexity for computing the threshold. Next, we consider the OFDMA MECO system, for which the optimal resource allocation is formulated as a mixed-integer problem. To solve this challenging problem and characterize its policy structure, a low-complexity sub-optimal algorithm is proposed by transforming the OFDMA problem to its TDMA counterpart. The corresponding resource allocation is derived by defining an average offloading priority function and shown to have close-to-optimal performance in simulation.

Experiment-driven Characterization of Full-Duplex Wireless Systems We present an experiment-based characterization of passive suppression and active self-interference cancellation mechanisms in full-duplex wireless communication systems. In particular, we consider passive suppression due to antenna separation at the same node, and active cancellation in analog and/or digital domain. First, we show that the average amount of cancellation increases for active cance...

Multi-stage genetic programming: A new strategy to nonlinear system modeling This paper presents a new multi-stage genetic programming (MSGP) strategy for modeling nonlinear systems. The proposed strategy is based on incorporating the individual effect of predictor variables and the interactions among them to provide more accurate simulations. According to the MSGP strategy, an efficient formulation for a problem comprises different terms. In the first stage of the MSGP-based analysis, the output variable is formulated in terms of an influencing variable. Thereafter, the error between the actual and the predicted value is formulated in terms of a new variable. Finally, the interaction term is derived by formulating the difference between the actual values and the values predicted by the individually developed terms. The capabilities of MSGP are illustrated by applying it to the formulation of different complex engineering problems. The problems analyzed herein include the following: (i) simulation of pH neutralization process, (ii) prediction of surface roughness in end milling, and (iii) classification of soil liquefaction conditions. The validity of the proposed strategy is confirmed by applying the derived models to the parts of the experimental results that were not included in the analyses. Further, the external validation of the models is verified using several statistical criteria recommended by other researchers. The MSGP-based solutions are capable of effectively simulating the nonlinear behavior of the investigated systems. The results of MSGP are found to be more accurate than those of standard GP and artificial neural network-based models.

Placing Virtual Machines to Optimize Cloud Gaming Experience Optimizing cloud gaming experience is no easy task due to the complex tradeoff between gamer quality of experience (QoE) and provider net profit. We tackle the challenge and study an optimization problem to maximize the cloud gaming provider's total profit while achieving just-good-enough QoE. We conduct measurement studies to derive the QoE and performance models. We formulate and optimally solve the problem. The optimization problem has exponential running time, and we develop an efficient heuristic algorithm. We also present an alternative formulation and algorithms for closed cloud gaming services with dedicated infrastructures, where the profit is not a concern and overall gaming QoE needs to be maximized. We present a prototype system and testbed using off-the-shelf virtualization software, to demonstrate the practicality and efficiency of our algorithms. Our experience on realizing the testbed sheds some lights on how cloud gaming providers may build up their own profitable services. Last, we conduct extensive trace-driven simulations to evaluate our proposed algorithms. The simulation results show that the proposed heuristic algorithms: (i) produce close-to-optimal solutions, (ii) scale to large cloud gaming services with 20,000 servers and 40,000 gamers, and (iii) outperform the state-of-the-art placement heuristic, e.g., by up to 3.5 times in terms of net profits.

Neural network adaptive tracking control for a class of uncertain switched nonlinear systems. •Study the method of the tracking control of the switched uncertain nonlinear systems under arbitrary switching signal controller.•A multilayer neural network adaptive controller with multilayer weight norm adaptive estimation is been designed.•The adaptive law is expand from calculation the second layer weight of neural network to both of the two layers weight.•The controller proposed improve the tracking error performance of the closed-loop system greatly.

Ethical Considerations Of Applying Robots In Kindergarten Settings: Towards An Approach From A Macroperspective In child-robot interaction (cHRI) research, many studies pursue the goal to develop interactive systems that can be applied in everyday settings. For early education, increasingly, the setting of a kindergarten is targeted. However, when cHRI and research are brought into a kindergarten, a range of ethical and related procedural aspects have to be considered and dealt with. While ethical models elaborated within other human-robot interaction settings, e.g., assisted living contexts, can provide some important indicators for relevant issues, we argue that it is important to start developing a systematic approach to identify and tackle those ethical issues which rise with cHRI in kindergarten settings on a more global level and address the impact of the technology from a macroperspective beyond the effects on the individual. Based on our experience in conducting studies with children in general and pedagogical considerations on the role of the institution of kindergarten in specific, in this paper, we enfold some relevant aspects that have barely been addressed in an explicit way in current cHRI research. Four areas are analyzed and key ethical issues are identified in each area: (1) the institutional setting of a kindergarten, (2) children as a vulnerable group, (3) the caregivers' role, and (4) pedagogical concepts. With our considerations, we aim at (i) broadening the methodology of the current studies within the area of cHRI, (ii) revalidate it based on our comprehensive empirical experience with research in kindergarten settings, both laboratory and real-world contexts, and (iii) provide a framework for the development of a more systematic approach to address the ethical issues in cHRI research within kindergarten settings.

Efficient Ladder-Style DenseNets for Semantic Segmentation of Large Images Recent progress of deep image classification models provides great potential for improving related computer vision tasks. However, the transition to semantic segmentation is hampered by strict memory limitations of contemporary GPUs. The extent of feature map caching required by convolutional backprop poses significant challenges even for moderately sized Pascal images, while requiring careful architectural considerations when input resolution is in the megapixel range. To address these concerns, we propose a novel ladder-style DenseNet-based architecture which features high modelling power, efficient upsampling, and inherent spatial efficiency which we unlock with checkpointing. The resulting models deliver high performance and allow training at megapixel resolution on commodity hardware. The presented experimental results outperform the state-of-the-art in terms of prediction accuracy and execution speed on Cityscapes, VOC 2012, CamVid and ROB 2018 datasets. Source code at https://github.com/ivankreso/LDN.

Generative Adversarial Networks for Parallel Transportation Systems. Generative Adversaria Networks (GANs) have emerged as a promising and effective mechanism for machine learning due to its recent successful applications. GANs share the same idea of producing, testing, acquiring, and utilizing data as well as knowledge based on artificial systems, computational experiments, and parallel execution of actual and virtual scenarios, as outlined in the theory of parall...

Deep Multi-Modal Object Detection and Semantic Segmentation for Autonomous Driving: Datasets, Methods, and Challenges AbstractRecent advancements in perception for autonomous driving are driven by deep learning. In order to achieve robust and accurate scene understanding, autonomous vehicles are usually equipped with different sensors (e.g. cameras, LiDARs, Radars), and multiple sensing modalities can be fused to exploit their complementary properties. In this context, many methods have been proposed for deep multi-modal perception problems. However, there is no general guideline for network architecture design, and questions of “what to fuse”, “when to fuse”, and “how to fuse” remain open. This review paper attempts to systematically summarize methodologies and discuss challenges for deep multi-modal object detection and semantic segmentation in autonomous driving. To this end, we first provide an overview of on-board sensors on test vehicles, open datasets, and background information for object detection and semantic segmentation in autonomous driving research. We then summarize the fusion methodologies and discuss challenges and open questions. In the appendix, we provide tables that summarize topics and methods. We also provide an interactive online platform to navigate each reference: https://boschresearch.github.io/multimodalperception/.

Enhanced Object Detection With Deep Convolutional Neural Networks for Advanced Driving Assistance Object detection is a critical problem for advanced driving assistance systems (ADAS). Recently, convolutional neural networks (CNN) achieved large successes on object detection, with performance improvement over traditional approaches, which use hand-engineered features. However, due to the challenging driving environment (e.g., large object scale variation, object occlusion, and bad light conditions), popular CNN detectors do not achieve very good object detection accuracy over the KITTI autonomous driving benchmark dataset. In this paper, we propose three enhancements for CNN-based visual object detection for ADAS. To address the large object scale variation challenge, deconvolution and fusion of CNN feature maps are proposed to add context and deeper features for better object detection at low feature map scales. In addition, soft non-maximal suppression (NMS) is applied across object proposals at different feature scales to address the object occlusion challenge. As the cars and pedestrians have distinct aspect ratio features, we measure their aspect ratio statistics and exploit them to set anchor boxes properly for better object matching and localization. The proposed CNN enhancements are evaluated with various image input sizes by experiments over KITTI dataset. The experimental results demonstrate the effectiveness of the proposed enhancements with good detection performance over KITTI test set.

MFR-CNN: Incorporating Multi-Scale Features and Global Information for Traffic Object Detection. Object detection plays an important role in intelligent transportation systems and intelligent vehicles. Although the topic of object detection has been studied for decades, it is still challenging to accurately detect objects under complex scenarios. The contributing factors for challenges include diversified object and background appearance, motion blur, adverse weather conditions, and complex i...

ParaUDA: Invariant Feature Learning With Auxiliary Synthetic Samples for Unsupervised Domain Adaptation Recognizing and locating objects by algorithms are essential and challenging issues for Intelligent Transportation Systems. However, the increasing demand for much labeled data hinders the further application of deep learning-based object detection. One of the optimal solutions is to train the target model with an existing dataset and then adapt it to new scenes, namely Unsupervised Domain Adaptation (UDA). However, most of existing methods at the pixel level mainly focus on adapting the model from source domain to target domain and ignore the essence of UDA to learn domain-invariant feature learning. Meanwhile, almost all methods at the feature level ignore to make conditional distributions matched for UDA while conducting feature alignment between source and target domain. Considering these problems, this paper proposes the ParaUDA, a novel framework of learning invariant representations for UDA in two aspects: pixel level and feature level. At the pixel level, we adopt CycleGAN to conduct domain transfer and convert the problem of original unsupervised domain adaptation to supervised domain adaptation. At the feature level, we adopt an adversarial adaption model to learn domain-invariant representation by aligning the distributions of domains between different image pairs with same mixture distributions. We evaluate our proposed framework in different scenes, from synthetic scenes to real scenes, from normal weather to challenging weather, and from scenes across cameras. The results of all the above experiments show that ParaUDA is effective and robust for adapting object detection models from source scenes to target scenes.

China's 12-Year Quest of Autonomous Vehicular Intelligence: The Intelligent Vehicles Future Challenge Program In this article, we introduce the Intelligent Vehicles Future Challenge of China (IVFC), which has lasted 12 years. Some key features of the tests and a few interesting findings of IVFC are selected and presented. Through the IVFCs held between 2009 and 2020, we gradually established a set of theories, methods, and tools to collect tests? data and efficiently evaluate the performance of autonomous vehicles so that we could learn how to improve both the autonomous vehicles and the testing system itself.

Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation Object detection performance, as measured on the canonical PASCAL VOC dataset, has plateaued in the last few years. The best-performing methods are complex ensemble systems that typically combine multiple low-level image features with high-level context. In this paper, we propose a simple and scalable detection algorithm that improves mean average precision (mAP) by more than 30% relative to the previous best result on VOC 2012 -- achieving a mAP of 53.3%. Our approach combines two key insights: (1) one can apply high-capacity convolutional neural networks (CNNs) to bottom-up region proposals in order to localize and segment objects and (2) when labeled training data is scarce, supervised pre-training for an auxiliary task, followed by domain-specific fine-tuning, yields a significant performance boost. Since we combine region proposals with CNNs, we call our method R-CNN: Regions with CNN features. We also present experiments that provide insight into what the network learns, revealing a rich hierarchy of image features. Source code for the complete system is available at http://www.cs.berkeley.edu/~rbg/rcnn.

A comparative study of texture measures with classification based on featured distributions This paper evaluates the performance both of some texture measures which have been successfully used in various applications and of some new promising approaches proposed recently. For classification a method based on Kullback discrimination of sample and prototype distributions is used. The classification results for single features with one-dimensional feature value distributions and for pairs of complementary features with two-dimensional distributions are presented

Social Perception and Steering for Online Avatars This paper presents work on a new platform for producing realistic group conversation dynamics in shared virtual environments. An avatar, representing users, should perceive the surrounding social environment just as humans would, and use the perceptual information for driving low level reactive behaviors. Unconscious reactions serve as evidence of life, and can also signal social availability and spatial awareness to others. These behaviors get lost when avatar locomotion requires explicit user control. For automating such behaviors we propose a steering layer in the avatars that manages a set of prioritized behaviors executed at different frequencies, which can be activated or deactivated and combined together. This approach gives us enough flexibility to model the group dynamics of social interactions as a set of social norms that activate relevant steering behaviors. A basic set of behaviors is described for conversations, some of which generate a social force field that makes the formation of conversation groups fluidly adapt to external and internal noise, through avatar repositioning and reorientations. The resulting social group behavior appears relatively robust, but perhaps more importantly, it starts to bring a new sense of relevance and continuity to the virtual bodies that often get separated from the ongoing conversation in the chat window.

Node Reclamation and Replacement for Long-Lived Sensor Networks When deployed for long-term tasks, the energy required to support sensor nodes' activities is far more than the energy that can be preloaded in their batteries. No matter how the battery energy is conserved, once the energy is used up, the network life terminates. Therefore, guaranteeing long-term energy supply has persisted as a big challenge. To address this problem, we propose a node reclamation and replacement (NRR) strategy, with which a mobile robot or human labor called mobile repairman (MR) periodically traverses the sensor network, reclaims nodes with low or no power supply, replaces them with fully charged ones, and brings the reclaimed nodes back to an energy station for recharging. To effectively and efficiently realize the strategy, we develop an adaptive rendezvous-based two-tier scheduling scheme (ARTS) to schedule the replacement/reclamation activities of the MR and the duty cycles of nodes. Extensive simulations have been conducted to verify the effectiveness and efficiency of the ARTS scheme.

Haptic feedback for enhancing realism of walking simulations. In this paper, we describe several experiments whose goal is to evaluate the role of plantar vibrotactile feedback in enhancing the realism of walking experiences in multimodal virtual environments. To achieve this goal we built an interactive and a noninteractive multimodal feedback system. While during the use of the interactive system subjects physically walked, during the use of the noninteractive system the locomotion was simulated while subjects were sitting on a chair. In both the configurations subjects were exposed to auditory and audio-visual stimuli presented with and without the haptic feedback. Results of the experiments provide a clear preference toward the simulations enhanced with haptic feedback showing that the haptic channel can lead to more realistic experiences in both interactive and noninteractive configurations. The majority of subjects clearly appreciated the added feedback. However, some subjects found the added feedback unpleasant. This might be due, on one hand, to the limits of the haptic simulation and, on the other hand, to the different individual desire to be involved in the simulations. Our findings can be applied to the context of physical navigation in multimodal virtual environments as well as to enhance the user experience of watching a movie or playing a video game.

Vehicular Sensing Networks in a Smart City: Principles, Technologies and Applications. Given the escalating population across the globe, it has become paramount to construct smart cities, aiming for improving the management of urban flows relying on efficient information and communication technologies (ICT). Vehicular sensing networks (VSNs) play a critical role in maintaining the efficient operation of smart cities. Naturally, there are numerous challenges to be solved before the w...

Dual-objective mixed integer linear program and memetic algorithm for an industrial group scheduling problem Group scheduling problems have attracted much attention owing to their many practical applications. This work proposes a new bi-objective serial-batch group scheduling problem considering the constraints of sequence-dependent setup time, release time, and due time. It is originated from an important industrial process, i.e., wire rod and bar rolling process in steel production systems. Two objecti...

Reduced-Order Observer Design for Switched Descriptor Systems With Unknown Inputs This paper presents a systematical reduced-order observer design method for a class of switched descriptor systems containing unknown inputs (UIs) in both the dynamic and the output equations. Generally speaking, when the output of the system contains UIs, the reduced-order observer design will become much more challenging. In order to overcome the difficulty brought by the UIs in the output, first, by introducing a new UI vector, a new equivalent system is obtained, which does not contain UIs in the corresponding output any more. Then, for the purpose of reduced-order observer design, the observer matching condition (OMC) and the minimal phase condition (MPC) are discussed, and it is shown that the new general system maintains both the OMC and the MPC. Subsequently, based on the discussions on the existence conditions, a reduced-order unknown input observer is developed to asymptotically estimate the state of the original system. Finally, an electronic circuit example is given to show the effectiveness of the proposed method.

The Sybil Attack Large-scale peer-to-peer systems facesecurity threats from faulty or hostile remotecomputing elements. To resist these threats, manysuch systems employ redundancy. However, if asingle faulty entity can present multiple identities,it can control a substantial fraction of the system,thereby undermining this redundancy. Oneapproach to preventing these &quot;Sybil attacks&quot; is tohave a trusted agency certify identities. Thispaper shows that, without a logically centralizedauthority, Sybil...

BLEU: a method for automatic evaluation of machine translation Human evaluations of machine translation are extensive but expensive. Human evaluations can take months to finish and involve human labor that can not be reused. We propose a method of automatic machine translation evaluation that is quick, inexpensive, and language-independent, that correlates highly with human evaluation, and that has little marginal cost per run. We present this method as an automated understudy to skilled human judges which substitutes for them when there is need for quick or frequent evaluations.

Computational thinking Summary form only given. My vision for the 21st century, Computational Thinking, will be a fundamental skill used by everyone in the world. To reading, writing, and arithmetic, we should add computational thinking to every child's analytical ability. Computational thinking involves solving problems, designing systems, and understanding human behavior by drawing on the concepts fundamental to computer science. Thinking like a computer scientist means more than being able to program a computer. It requires the ability to abstract and thus to think at multiple levels of abstraction. In this talk I will give many examples of computational thinking, argue that it has already influenced other disciplines, and promote the idea that teaching computational thinking can not only inspire future generations to enter the field of computer science but benefit people in all fields.

Fuzzy logic in control systems: fuzzy logic controller. I.

Switching between stabilizing controllers This paper deals with the problem of switching between several linear time-invariant (LTI) controllers—all of them capable of stabilizing a speci4c LTI process—in such a way that the stability of the closed-loop system is guaranteed for any switching sequence. We show that it is possible to 4nd realizations for any given family of controller transfer matrices so that the closed-loop system remains stable, no matter how we switch among the controller. The motivation for this problem is the control of complex systems where con8icting requirements make a single LTI controller unsuitable. ? 2002 Published by Elsevier Science Ltd.

Tabu Search - Part I

Bidirectional recurrent neural networks In the first part of this paper, a regular recurrent neural network (RNN) is extended to a bidirectional recurrent neural network (BRNN). The BRNN can be trained without the limitation of using input information just up to a preset future frame. This is accomplished by training it simultaneously in positive and negative time direction. Structure and training procedure of the proposed network are explained. In regression and classification experiments on artificial data, the proposed structure gives better results than other approaches. For real data, classification experiments for phonemes from the TIMIT database show the same tendency. In the second part of this paper, it is shown how the proposed bidirectional structure can be easily modified to allow efficient estimation of the conditional posterior probability of complete symbol sequences without making any explicit assumption about the shape of the distribution. For this part, experiments on real data are reported

An intensive survey of fair non-repudiation protocols With the phenomenal growth of the Internet and open networks in general, security services, such as non-repudiation, become crucial to many applications. Non-repudiation services must ensure that when Alice sends some information to Bob over a network, neither Alice nor Bob can deny having participated in a part or the whole of this communication. Therefore a fair non-repudiation protocol has to generate non-repudiation of origin evidences intended to Bob, and non-repudiation of receipt evidences destined to Alice. In this paper, we clearly define the properties a fair non-repudiation protocol must respect, and give a survey of the most important non-repudiation protocols without and with trusted third party (TTP). For the later ones we discuss the evolution of the TTP's involvement and, between others, describe the most recent protocol using a transparent TTP. We also discuss some ad-hoc problems related to the management of non-repudiation evidences.

Dynamic movement and positioning of embodied agents in multiparty conversations For embodied agents to engage in realistic multiparty conversation, they must stand in appropriate places with respect to other agents and the environment. When these factors change, such as an agent joining the conversation, the agents must dynamically move to a new location and/or orientation to accommodate. This paper presents an algorithm for simulating movement of agents based on observed human behavior using techniques developed for pedestrian movement in crowd simulations. We extend a previous group conversation simulation to include an agent motion algorithm. We examine several test cases and show how the simulation generates results that mirror real-life conversation settings.

An improved genetic algorithm with conditional genetic operators and its application to set-covering problem The genetic algorithm (GA) is a popular, biologically inspired optimization method. However, in the GA there is no rule of thumb to design the GA operators and select GA parameters. Instead, trial-and-error has to be applied. In this paper we present an improved genetic algorithm in which crossover and mutation are performed conditionally instead of probability. Because there are no crossover rate and mutation rate to be selected, the proposed improved GA can be more easily applied to a problem than the conventional genetic algorithms. The proposed improved genetic algorithm is applied to solve the set-covering problem. Experimental studies show that the improved GA produces better results over the conventional one and other methods.

Lane-level traffic estimations using microscopic traffic variables This paper proposes a novel inference method to estimate lane-level traffic flow, time occupancy and vehicle inter-arrival time on road segments where local information could not be measured and assessed directly. The main contributions of the proposed method are 1) the ability to perform lane-level estimations of traffic flow, time occupancy and vehicle inter-arrival time and 2) the ability to adapt to different traffic regimes by assessing only microscopic traffic variables. We propose a modified Kriging estimation model which explicitly takes into account both spatial and temporal variability. Performance evaluations are conducted using real-world data under different traffic regimes and it is shown that the proposed method outperforms a Kalman filter-based approach.

Convolutional Neural Network-Based Classification of Driver's Emotion during Aggressive and Smooth Driving Using Multi-Modal Camera Sensors. Because aggressive driving often causes large-scale loss of life and property, techniques for advance detection of adverse driver emotional states have become important for the prevention of aggressive driving behaviors. Previous studies have primarily focused on systems for detecting aggressive driver emotion via smart-phone accelerometers and gyro-sensors, or they focused on methods of detecting physiological signals using electroencephalography (EEG) or electrocardiogram (ECG) sensors. Because EEG and ECG sensors cause discomfort to drivers and can be detached from the driver's body, it becomes difficult to focus on bio-signals to determine their emotional state. Gyro-sensors and accelerometers depend on the performance of GPS receivers and cannot be used in areas where GPS signals are blocked. Moreover, if driving on a mountain road with many quick turns, a driver's emotional state can easily be misrecognized as that of an aggressive driver. To resolve these problems, we propose a convolutional neural network (CNN)-based method of detecting emotion to identify aggressive driving using input images of the driver's face, obtained using near-infrared (NIR) light and thermal camera sensors. In this research, we conducted an experiment using our own database, which provides a high classification accuracy for detecting driver emotion leading to either aggressive or smooth (i.e., relaxed) driving. Our proposed method demonstrates better performance than existing methods.

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

Optimization insisted watermarking model: hybrid firefly and Jaya algorithm for video copyright protection In this digital era, illegitimate redistribution and protection of digital multimedia have to turn out to be the critical issue. Therefore, digital watermarking has been introduced for avoiding illegitimate works and also to ensure security and authentication as well. "Digital video watermarking is a method to hide some kind of data like audio, image, text into digital video sequences which is nothing but orders of successive still images." This paper intends to formulate a novel video watermarking framework that includes three stages, such as (i) optimal video frame prediction, (ii) watermark embedding process and (iii) watermark extraction process. Very first, the optimal frames are determined using a new hybrid algorithm termed trial-based update on Jaya plus firefly (TU-JF) algorithm, in such a way that the peak signal-to-noise ratio (PSNR) should be maximal. The frames are assigned with a label of one or zero, where label one denotes the frame with better PSNR (can select for embedding process) and label zero denotes the frame with reduced PSNR (cannot be used for embedding). Consequently, a data library is formed from the obtained results, where each frame of videos is determined with their gray-level co-occurrence matrix (GLCM) features and labels (can embed or not). As in the proposed model, the optimal frame prediction is carried out using deep belief network (DBN) framework; the obtained data are then trained in the model. The optimal frames could be predicted in an efficient manner while testing. The significant contribution of this work concerns the optimization of hidden neurons in the DBN framework, which helps to enhance prediction accuracy. At last, the "watermark embedding process" and "watermark extraction process" are done by which the image could be embedded within the optimal frames.

Multi-stage genetic programming: A new strategy to nonlinear system modeling This paper presents a new multi-stage genetic programming (MSGP) strategy for modeling nonlinear systems. The proposed strategy is based on incorporating the individual effect of predictor variables and the interactions among them to provide more accurate simulations. According to the MSGP strategy, an efficient formulation for a problem comprises different terms. In the first stage of the MSGP-based analysis, the output variable is formulated in terms of an influencing variable. Thereafter, the error between the actual and the predicted value is formulated in terms of a new variable. Finally, the interaction term is derived by formulating the difference between the actual values and the values predicted by the individually developed terms. The capabilities of MSGP are illustrated by applying it to the formulation of different complex engineering problems. The problems analyzed herein include the following: (i) simulation of pH neutralization process, (ii) prediction of surface roughness in end milling, and (iii) classification of soil liquefaction conditions. The validity of the proposed strategy is confirmed by applying the derived models to the parts of the experimental results that were not included in the analyses. Further, the external validation of the models is verified using several statistical criteria recommended by other researchers. The MSGP-based solutions are capable of effectively simulating the nonlinear behavior of the investigated systems. The results of MSGP are found to be more accurate than those of standard GP and artificial neural network-based models.

An improved genetic algorithm with conditional genetic operators and its application to set-covering problem The genetic algorithm (GA) is a popular, biologically inspired optimization method. However, in the GA there is no rule of thumb to design the GA operators and select GA parameters. Instead, trial-and-error has to be applied. In this paper we present an improved genetic algorithm in which crossover and mutation are performed conditionally instead of probability. Because there are no crossover rate and mutation rate to be selected, the proposed improved GA can be more easily applied to a problem than the conventional genetic algorithms. The proposed improved genetic algorithm is applied to solve the set-covering problem. Experimental studies show that the improved GA produces better results over the conventional one and other methods.

Hybrid Whale Optimization Algorithm with simulated annealing for feature selection. •Four hybrid feature selection methods for classification task are proposed.•Our hybrid method combines Whale Optimization Algorithm with simulated annealing.•Eighteen UCI datasets were used in the experiments.•Our approaches result a higher accuracy by using less number of features.

Solving the dynamic weapon target assignment problem by an improved artificial bee colony algorithm with heuristic factor initialization. •Put forward an improved artificial bee colony algorithm based on ranking selection and elite guidance.•Put forward 4 rule-based heuristic factors: Wc, Rc, TRc and TRcL.•The heuristic factors are used in population initialization to improve the quality of the initial solutions in DWTA solving.•The heuristic factor initialization method is combined with the improved ABC algorithm to solve the DWTA problem.

A monarch butterfly optimization-based neural network simulator for prediction of siro-spun yarn tenacity Yarn tenacity directly affects the winding and knitting efficiency as well as warp and weft breakages during weaving process and therefore, is considered as the most important parameter to be controlled during yarn spinning process. Yarn tenacity is dependent on fiber properties and process parameters. Exploring the relationship between fiber properties, process parameters and yarn tenacity is very important to optimize the selection of raw materials and improve yarn quality. In this study, an efficient monarch butterfly optimization-based neural network simulator called MBONN was developed to predict the tenacity of siro-spun yarns from some process parameters and fiber properties. To this end, an experimental dataset was obtained with fiber fineness, yarn twist factor, yarn linear density and strand spacing as the input variables and yarn tenacity as the output parameter. In the proposed MBONN, a monarch butterfly optimization algorithm is applied as a global search method to evolve weights of a multilayer perception (MLP) neural network. The prediction accuracy of the MBONN was compared with that of a MLP neural network trained with back propagation algorithm, MLP neural network trained with genetic algorithms and linear regression model. The results indicated that the prediction accuracy of the proposed MBONN is statistically superior to that of other models. The effect of fiber fineness, yarn linear density, twist factor and strand spacing on yarn tenacity was investigated using the proposed MBONN. Additionally, the observed trends in variation of yarn tenacity with fiber and process parameters were discussed with reference to the yarn internal structure. It was established that higher migration parameters result in increasing the siro-spun yarn tenacity. It was found that the yarns with higher migration parameters benefit from a more coherent self-locking structure which severely restricts fiber slippage, thereby increasing the yarn tenacity.

An improved artificial bee colony algorithm for balancing local and global search behaviors in continuous optimization The artificial bee colony, ABC for short, algorithm is population-based iterative optimization algorithm proposed for solving the optimization problems with continuously-structured solution space. Although ABC has been equipped with powerful global search capability, this capability can cause poor intensification on found solutions and slow convergence problem. The occurrence of these issues is originated from the search equations proposed for employed and onlooker bees, which only updates one decision variable at each trial. In order to address these drawbacks of the basic ABC algorithm, we introduce six search equations for the algorithm and three of them are used by employed bees and the rest of equations are used by onlooker bees. Moreover, each onlooker agent can modify three dimensions or decision variables of a food source at each attempt, which represents a possible solution for the optimization problems. The proposed variant of ABC algorithm is applied to solve basic, CEC2005, CEC2014 and CEC2015 benchmark functions. The obtained results are compared with results of the state-of-art variants of the basic ABC algorithm, artificial algae algorithm, particle swarm optimization algorithm and its variants, gravitation search algorithm and its variants and etc. Comparisons are conducted for measurement of the solution quality, robustness and convergence characteristics of the algorithms. The obtained results and comparisons show the experimentally validation of the proposed ABC variant and success in solving the continuous optimization problems dealt with the study.

On the security of public key protocols Recently the use of public key encryption to provide secure network communication has received considerable attention. Such public key systems are usually effective against passive eavesdroppers, who merely tap the lines and try to decipher the message. It has been pointed out, however, that an improperly designed protocol could be vulnerable to an active saboteur, one who may impersonate another user or alter the message being transmitted. Several models are formulated in which the security of protocols can be discussed precisely. Algorithms and characterizations that can be used to determine protocol security in these models are given.

QoE-Driven Edge Caching in Vehicle Networks Based on Deep Reinforcement Learning The Internet of vehicles (IoV) is a large information interaction network that collects information on vehicles, roads and pedestrians. One of the important uses of vehicle networks is to meet the entertainment needs of driving users through communication between vehicles and roadside units (RSUs). Due to the limited storage space of RSUs, determining the content cached in each RSU is a key challenge. With the development of 5G and video editing technology, short video systems have become increasingly popular. Current widely used cache update methods, such as partial file precaching and content popularity- and user interest-based determination, are inefficient for such systems. To solve this problem, this paper proposes a QoE-driven edge caching method for the IoV based on deep reinforcement learning. First, a class-based user interest model is established. Compared with the traditional file popularity- and user interest distribution-based cache update methods, the proposed method is more suitable for systems with a large number of small files. Second, a quality of experience (QoE)-driven RSU cache model is established based on the proposed class-based user interest model. Third, a deep reinforcement learning method is designed to address the QoE-driven RSU cache update issue effectively. The experimental results verify the effectiveness of the proposed algorithm.

Image information and visual quality Measurement of visual quality is of fundamental importance to numerous image and video processing applications. The goal of quality assessment (QA) research is to design algorithms that can automatically assess the quality of images or videos in a perceptually consistent manner. Image QA algorithms generally interpret image quality as fidelity or similarity with a "reference" or "perfect" image in some perceptual space. Such "full-reference" QA methods attempt to achieve consistency in quality prediction by modeling salient physiological and psychovisual features of the human visual system (HVS), or by signal fidelity measures. In this paper, we approach the image QA problem as an information fidelity problem. Specifically, we propose to quantify the loss of image information to the distortion process and explore the relationship between image information and visual quality. QA systems are invariably involved with judging the visual quality of "natural" images and videos that are meant for "human consumption." Researchers have developed sophisticated models to capture the statistics of such natural signals. Using these models, we previously presented an information fidelity criterion for image QA that related image quality with the amount of information shared between a reference and a distorted image. In this paper, we propose an image information measure that quantifies the information that is present in the reference image and how much of this reference information can be extracted from the distorted image. Combining these two quantities, we propose a visual information fidelity measure for image QA. We validate the performance of our algorithm with an extensive subjective study involving 779 images and show that our method outperforms recent state-of-the-art image QA algorithms by a sizeable margin in our simulations. The code and the data from the subjective study are available at the LIVE website.

Stabilization of switched continuous-time systems with all modes unstable via dwell time switching Stabilization of switched systems composed fully of unstable subsystems is one of the most challenging problems in the field of switched systems. In this brief paper, a sufficient condition ensuring the asymptotic stability of switched continuous-time systems with all modes unstable is proposed. The main idea is to exploit the stabilization property of switching behaviors to compensate the state divergence made by unstable modes. Then, by using a discretized Lyapunov function approach, a computable sufficient condition for switched linear systems is proposed in the framework of dwell time; it is shown that the time intervals between two successive switching instants are required to be confined by a pair of upper and lower bounds to guarantee the asymptotic stability. Based on derived results, an algorithm is proposed to compute the stability region of admissible dwell time. A numerical example is proposed to illustrate our approach.

Software-Defined Networking: A Comprehensive Survey The Internet has led to the creation of a digital society, where (almost) everything is connected and is accessible from anywhere. However, despite their widespread adoption, traditional IP networks are complex and very hard to manage. It is both difficult to configure the network according to predefined policies, and to reconfigure it to respond to faults, load, and changes. To make matters even more difficult, current networks are also vertically integrated: the control and data planes are bundled together. Software-defined networking (SDN) is an emerging paradigm that promises to change this state of affairs, by breaking vertical integration, separating the network's control logic from the underlying routers and switches, promoting (logical) centralization of network control, and introducing the ability to program the network. The separation of concerns, introduced between the definition of network policies, their implementation in switching hardware, and the forwarding of traffic, is key to the desired flexibility: by breaking the network control problem into tractable pieces, SDN makes it easier to create and introduce new abstractions in networking, simplifying network management and facilitating network evolution. In this paper, we present a comprehensive survey on SDN. We start by introducing the motivation for SDN, explain its main concepts and how it differs from traditional networking, its roots, and the standardization activities regarding this novel paradigm. Next, we present the key building blocks of an SDN infrastructure using a bottom-up, layered approach. We provide an in-depth analysis of the hardware infrastructure, southbound and northbound application programming interfaces (APIs), network virtualization layers, network operating systems (SDN controllers), network programming languages, and network applications. We also look at cross-layer problems such as debugging and troubleshooting. In an effort to anticipate the future evolution of this - ew paradigm, we discuss the main ongoing research efforts and challenges of SDN. In particular, we address the design of switches and control platforms - with a focus on aspects such as resiliency, scalability, performance, security, and dependability - as well as new opportunities for carrier transport networks and cloud providers. Last but not least, we analyze the position of SDN as a key enabler of a software-defined environment.

An ID-Based Linearly Homomorphic Signature Scheme and Its Application in Blockchain. Identity-based cryptosystems mean that public keys can be directly derived from user identifiers, such as telephone numbers, email addresses, and social insurance number, and so on. So they can simplify key management procedures of certificate-based public key infrastructures and can be used to realize authentication in blockchain. Linearly homomorphic signature schemes allow to perform linear computations on authenticated data. And the correctness of the computation can be publicly verified. Although a series of homomorphic signature schemes have been designed recently, there are few homomorphic signature schemes designed in identity-based cryptography. In this paper, we construct a new ID-based linear homomorphic signature scheme, which avoids the shortcomings of the use of public-key certificates. The scheme is proved secure against existential forgery on adaptively chosen message and ID attack under the random oracle model. The ID-based linearly homomorphic signature schemes can be applied in e-business and cloud computing. Finally, we show how to apply it to realize authentication in blockchain.

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

Secure message classification services through identity-based signcryption with equality test towards the Internet of vehicles To provide a classification function that has the ability to promote the management of signcrypted messages transmitted by the numerous vehicles in the IoV system, in this paper, we construct an identity-based signcryption with equality test scheme (IBSC-ET) for the first time. Our scheme not only ensures the integrity, confidentiality as well as authentication of messages, but also enables the cloud server to execute the equality test between two ciphertexts signcrypted by the same or different public keys to decide whether the same plaintext is contained, which is counted as the essential factor contributing to the classification. Besides, the identity-based mechanism greatly improves the efficiency since the certificate management problem is avoided. Furthermore, the security of the introduced scheme can be proved in the random oracle model under the Computational Diffie-Hellman Assumption (CDHA) as well as the Bilinear Diffie-Hellman Assumption (BDHA). In the end, the feasibility and efficiency of our proposed are demonstrated through the performance analysis.

Reliable Computation Offloading for Edge-Computing-Enabled Software-Defined IoV Internet of Vehicles (IoV) has drawn great interest recent years. Various IoV applications have emerged for improving the safety, efficiency, and comfort on the road. Cloud computing constitutes a popular technique for supporting delay-tolerant entertainment applications. However, for advanced latency-sensitive applications (e.g., auto/assisted driving and emergency failure management), cloud computing may result in excessive delay. Edge computing, which extends computing and storage capabilities to the edge of the network, emerges as an attractive technology. Therefore, to support these computationally intensive and latency-sensitive applications in IoVs, in this article, we integrate mobile-edge computing nodes (i.e., mobile vehicles) and fixed edge computing nodes (i.e., fixed road infrastructures) to provide low-latency computing services cooperatively. For better exploiting these heterogeneous edge computing resources, the concept of software-defined networking (SDN) and edge-computing-aided IoV (EC-SDIoV) is conceived. Moreover, in a complex and dynamic IoV environment, the outage of both processing nodes and communication links becomes inevitable, which may have life-threatening consequences. In order to ensure the completion with high reliability of latency-sensitive IoV services, we introduce both partial computation offloading and reliable task allocation with the reprocessing mechanism to EC-SDIoV. Since the optimization problem is nonconvex and NP-hard, a heuristic algorithm, fault-tolerant particle swarm optimization algorithm is designed for maximizing the reliability (FPSO-MR) with latency constraints. Performance evaluation results validate that the proposed scheme is indeed capable of reducing the latency as well as improving the reliability of the EC-SDIoV.

Energy Efficiency Resource Allocation For D2d Communication Network Based On Relay Selection In order to solve the problem of spectrum resource shortage and energy consumption, we put forward a new model that combines with D2D communication and energy harvesting technology: energy harvesting-aided D2D communication network under the cognitive radio (EHA-CRD), where the D2D users harvest energy from the base station and the D2D source communicate with D2D destination by D2D relays. Our goals are to investigate the maximization energy efficiency (EE) of the network by joint time allocation and relay selection while taking into the constraints of the signal-to-noise ratio of D2D and the rates of the Cellular users. During this process, the energy collection time and communication time are randomly allocated. The maximization problem of EE can be divided into two sub-problems: (1) relay selection problem; (2) time optimization problem. For the first sub-problem, we propose a weighted sum maximum algorithm to select the best relay. For the last sub-problem, the EE maximization problem is non-convex problem with time. Thus, by using fractional programming theory, we transform it into a standard convex optimization problem, and we propose the optimization iterative algorithm to solve the convex optimization problem for obtaining the optimal solution. And, the simulation results show that the proposed relay selection algorithm and time optimization algorithm are significantly improved compared with the existing algorithms.

Repeated Game Analysis for Cooperative MAC With Incentive Design for Wireless Networks. Cooperative communications offer appealing potentials to improve quality of service (QoS) for wireless networks. Many existing works on cooperative communications assume that participation in cooperative relaying is unconditional. In practice, however, due to resource consumption, it is vital to provide incentives for selfish cooperating peer nodes. In this paper, we analyze a cooperative medium a...

Internet of Vehicles: Sensing-Aided Transportation Information Collection and Diffusion. In view of the emergence and rapid development of the Internet of Vehicles (IoV) and cloud computing, intelligent transport systems are beneficial in terms of enhancing the quality and interactivity of urban transportation services, reducing costs and resource wastage, and improving the traffic management capability. Efficient traffic management relies on the accurate and prompt acquisition as wel...

Blockchain-based secured event-information sharing protocol in internet of vehicles for smart cities An intelligent transportation system is an advanced application that aims to provide traffic congestion information, road accidents, emergency information to other vehicles, etc. In the traditional transportation systems, the road-side unit (RSU) acts as a central authority and plays an important role to maintain everything in the networks. To solve single point of failure, as well as to achieve data immutability, blockchain-based decentralized vehicular ad-hoc networks is very essential and important. In order to share critical information among all the road-side units, we design a blockchain-based decentralized vehicular ad-hoc network that supports data immutability property. We have also designed an authentication protocol for the vehicle’s user and a consensus mechanism to validate transactions. Besides, a smart contract mechanism has also been proposed. Informal security analysis confirms anonymity property with other security requirements. We have also calculated and discussed communication, computation, and storage overhead of the proposed scheme.

The Sybil Attack Large-scale peer-to-peer systems facesecurity threats from faulty or hostile remotecomputing elements. To resist these threats, manysuch systems employ redundancy. However, if asingle faulty entity can present multiple identities,it can control a substantial fraction of the system,thereby undermining this redundancy. Oneapproach to preventing these &quot;Sybil attacks&quot; is tohave a trusted agency certify identities. Thispaper shows that, without a logically centralizedauthority, Sybil...

Sequence to Sequence Learning with Neural Networks. Deep Neural Networks (DNNs) are powerful models that have achieved excellent performance on difficult learning tasks. Although DNNs work well whenever large labeled training sets are available, they cannot be used to map sequences to sequences. In this paper, we present a general end-to-end approach to sequence learning that makes minimal assumptions on the sequence structure. Our method uses a multilayered Long Short-Term Memory (LSTM) to map the input sequence to a vector of a fixed dimensionality, and then another deep LSTM to decode the target sequence from the vector. Our main result is that on an English to French translation task from the WMT-14 dataset, the translations produced by the LSTM achieve a BLEU score of 34.8 on the entire test set, where the LSTM's BLEU score was penalized on out-of-vocabulary words. Additionally, the LSTM did not have difficulty on long sentences. For comparison, a phrase-based SMT system achieves a BLEU score of 33.3 on the same dataset. When we used the LSTM to rerank the 1000 hypotheses produced by the aforementioned SMT system, its BLEU score increases to 36.5, which is close to the previous state of the art. The LSTM also learned sensible phrase and sentence representations that are sensitive to word order and are relatively invariant to the active and the passive voice. Finally, we found that reversing the order of the words in all source sentences (but not target sentences) improved the LSTM's performance markedly, because doing so introduced many short term dependencies between the source and the target sentence which made the optimization problem easier.

Chimp optimization algorithm. •A novel optimizer called Chimp Optimization Algorithm (ChOA) is proposed.•ChOA is inspired by individual intelligence and sexual motivation of chimps.•ChOA alleviates the problems of slow convergence rate and trapping in local optima.•The four main steps of Chimp hunting are implemented.

A bayesian network approach to traffic flow forecasting A new approach based on Bayesian networks for traffic flow forecasting is proposed. In this paper, traffic flows among adjacent road links in a transportation network are modeled as a Bayesian network. The joint probability distribution between the cause nodes (data utilized for forecasting) and the effect node (data to be forecasted) in a constructed Bayesian network is described as a Gaussian mixture model (GMM) whose parameters are estimated via the competitive expectation maximization (CEM) algorithm. Finally, traffic flow forecasting is performed under the criterion of minimum mean square error (mmse). The approach departs from many existing traffic flow forecasting models in that it explicitly includes information from adjacent road links to analyze the trends of the current link statistically. Furthermore, it also encompasses the issue of traffic flow forecasting when incomplete data exist. Comprehensive experiments on urban vehicular traffic flow data of Beijing and comparisons with several other methods show that the Bayesian network is a very promising and effective approach for traffic flow modeling and forecasting, both for complete data and incomplete data

A simplified dual neural network for quadratic programming with its KWTA application. The design, analysis, and application of a new recurrent neural network for quadratic programming, called simplified dual neural network, are discussed. The analysis mainly concentrates on the convergence property and the computational complexity of the neural network. The simplified dual neural network is shown to be globally convergent to the exact optimal solution. The complexity of the neural network architecture is reduced with the number of neurons equal to the number of inequality constraints. Its application to k-winners-take-all (KWTA) operation is discussed to demonstrate how to solve problems with this neural network.

NETWRAP: An NDN Based Real-TimeWireless Recharging Framework for Wireless Sensor Networks Using vehicles equipped with wireless energy transmission technology to recharge sensor nodes over the air is a game-changer for traditional wireless sensor networks. The recharging policy regarding when to recharge which sensor nodes critically impacts the network performance. So far only a few works have studied such recharging policy for the case of using a single vehicle. In this paper, we propose NETWRAP, an N DN based Real Time Wireless Rech arging Protocol for dynamic wireless recharging in sensor networks. The real-time recharging framework supports single or multiple mobile vehicles. Employing multiple mobile vehicles provides more scalability and robustness. To efficiently deliver sensor energy status information to vehicles in real-time, we leverage concepts and mechanisms from named data networking (NDN) and design energy monitoring and reporting protocols. We derive theoretical results on the energy neutral condition and the minimum number of mobile vehicles required for perpetual network operations. Then we study how to minimize the total traveling cost of vehicles while guaranteeing all the sensor nodes can be recharged before their batteries deplete. We formulate the recharge optimization problem into a Multiple Traveling Salesman Problem with Deadlines (m-TSP with Deadlines), which is NP-hard. To accommodate the dynamic nature of node energy conditions with low overhead, we present an algorithm that selects the node with the minimum weighted sum of traveling time and residual lifetime. Our scheme not only improves network scalability but also ensures the perpetual operation of networks. Extensive simulation results demonstrate the effectiveness and efficiency of the proposed design. The results also validate the correctness of the theoretical analysis and show significant improvements that cut the number of nonfunctional nodes by half compared to the static scheme while maintaining the network overhead at the same level.

Gender Bias in Coreference Resolution. We present an empirical study of gender bias in coreference resolution systems. We first introduce a novel, Winograd schema-style set of minimal pair sentences that differ only by pronoun gender. With these Winogender schemas, we evaluate and confirm systematic gender bias in three publicly-available coreference resolution systems, and correlate this bias with real-world and textual gender statistics.

Sustainable and Efficient Data Collection from WSNs to Cloud. The development of cloud computing pours great vitality into traditional wireless sensor networks (WSNs). The integration of WSNs and cloud computing has received a lot of attention from both academia and industry. However, collecting data from WSNs to cloud is not sustainable. Due to the weak communication ability of WSNs, uploading big sensed data to the cloud within the limited time becomes a b...

Diversity in Machine Learning. Machine learning methods have achieved good performance and been widely applied in various real-world applications. They can learn the model adaptively and be better fit for special requirements of different tasks. Generally, a good machine learning system is composed of plentiful training data, a good model training process, and an accurate inference. Many factors can affect the performance of the machine learning process, among which the diversity of the machine learning process is an important one. The diversity can help each procedure to guarantee a totally good machine learning: diversity of the training data ensures that the training data can provide more discriminative information for the model, diversity of the learned model (diversity in parameters of each model or diversity among different base models) makes each parameter/model capture unique or complement information and the diversity in inference can provide multiple choices each of which corresponds to a specific plausible local optimal result. Even though diversity plays an important role in the machine learning process, there is no systematical analysis of the diversification in the machine learning system. In this paper, we systematically summarize the methods to make data diversification, model diversification, and inference diversification in the machine learning process. In addition, the typical applications where the diversity technology improved the machine learning performance have been surveyed including the remote sensing imaging tasks, machine translation, camera relocalization, image segmentation, object detection, topic modeling, and others. Finally, we discuss some challenges of the diversity technology in machine learning and point out some directions in future work. Our analysis provides a deeper understanding of the diversity technology in machine learning tasks and hence can help design and learn more effective models for real-world applications.

Contact personalization using a score understanding method This paper presents a method to interpret the output of a classification (or regression) model. The interpretation is based on two concepts: the variable importance and the value importance of the variable. Unlike most of the state of art interpretation methods, our approach allows the interpretation of the model output for every instance. Understanding the score given by a model for one instance can for example lead to an immediate decision in a customer relational management (CRM) system. Moreover the proposed method does not depend on a particular model and is therefore usable for any model or software used to produce the scores.

Image saliency: From intrinsic to extrinsic context We propose a novel framework for automatic saliency estimation in natural images. We consider saliency to be an anomaly with respect to a given context that can be global or local. In the case of global context, we estimate saliency in the whole image relative to a large dictionary of images. Unlike in some prior methods, this dictionary is not annotated, i.e., saliency is assumed unknown. In the case of local context, we partition the image into patches and estimate saliency in each patch relative to a large dictionary of un-annotated patches from the rest of the image. We propose a unified framework that applies to both cases in three steps. First, given an input (image or patch) we extract k nearest neighbors from the dictionary. Then, we geometrically warp each neighbor to match the input. Finally, we derive the saliency map from the mean absolute error between the input and all its warped neighbors. This algorithm is not only easy to implement but also outperforms state-of-the-art methods.

Scatter Component Analysis: A Unified Framework for Domain Adaptation and Domain Generalization. This paper addresses classification tasks on a particular target domain in which labeled training data are only available from source domains different from (but related to) the target. Two closely related frameworks, domain adaptation and domain generalization, are concerned with such tasks, where the only difference between those frameworks is the availability of the unlabeled target data: domai...

Explaining Classifications For Individual Instances We present a method for explaining predictions for individual instances. The presented approach is general and can be used with all classification models that output probabilities. It is based on the decomposition of a model's predictions on individual contributions of each attribute. Our method works for the so-called black box models such as support vector machines, neural networks, and nearest neighbor algorithms, as well as for ensemble methods such as boosting and random forests. We demonstrate that the generated explanations closely follow the learned models and present a visualization technique that shows the utility of our approach and enables the comparison of different prediction methods.

ImageNet Classification with Deep Convolutional Neural Networks. We trained a large, deep convolutional neural network to classify the 1.2 million high-resolution images in the ImageNet LSVRC-2010 contest into the 1000 different classes. On the test data, we achieved top-1 and top-5 error rates of 37.5% and 17.0%, respectively, which is considerably better than the previous state-of-the-art. The neural network, which has 60 million parameters and 650,000 neurons, consists of five convolutional layers, some of which are followed by max-pooling layers, and three fully connected layers with a final 1000-way softmax. To make training faster, we used non-saturating neurons and a very efficient GPU implementation of the convolution operation. To reduce overfitting in the fully connected layers we employed a recently developed regularization method called \"dropout\" that proved to be very effective. We also entered a variant of this model in the ILSVRC-2012 competition and achieved a winning top-5 test error rate of 15.3%, compared to 26.2% achieved by the second-best entry.

Accurate Self-Localization in RFID Tag Information Grids Using FIR Filtering Grid navigation spaces nested with the radio-frequency identification (RFID) tags are promising for industrial and other needs, because each tag can deliver information about a local two-dimensional or three-dimensional surrounding. The approach, however, requires high accuracy in vehicle self-localization. Otherwise, errors may lead to collisions; possibly even fatal. We propose a new extended finite impulse response (EFIR) filtering algorithm and show that it meets this need. The EFIR filter requires an optimal averaging interval, but does not involve the noise statistics which are often not well known to the engineer. It is more accurate than the extended Kalman filter (EKF) under real operation conditions and its iterative algorithm has the Kalman form. Better performance of the proposed EFIR filter is demonstrated based on extensive simulations in a comparison to EKF, which is widely used in RFID tag grids. We also show that errors in noise covariances may provoke divergence in EKF, whereas the EFIR filter remains stable and is thus more robust.

Gradient-Based Learning Applied to Document Recognition Multilayer neural networks trained with the back-propagation algorithm constitute the best example of a successful gradient based learning technique. Given an appropriate network architecture, gradient-based learning algorithms can be used to synthesize a complex decision surface that can classify high-dimensional patterns, such as handwritten characters, with minimal preprocessing. This paper rev...

Latent dirichlet allocation We describe latent Dirichlet allocation (LDA), a generative probabilistic model for collections of discrete data such as text corpora. LDA is a three-level hierarchical Bayesian model, in which each item of a collection is modeled as a finite mixture over an underlying set of topics. Each topic is, in turn, modeled as an infinite mixture over an underlying set of topic probabilities. In the context of text modeling, the topic probabilities provide an explicit representation of a document. We present efficient approximate inference techniques based on variational methods and an EM algorithm for empirical Bayes parameter estimation. We report results in document modeling, text classification, and collaborative filtering, comparing to a mixture of unigrams model and the probabilistic LSI model.

A study on the use of non-parametric tests for analyzing the evolutionary algorithms' behaviour: a case study on the CEC'2005 Special Session on Real Parameter Optimization In recent years, there has been a growing interest for the experimental analysis in the field of evolutionary algorithms. It is noticeable due to the existence of numerous papers which analyze and propose different types of problems, such as the basis for experimental comparisons of algorithms, proposals of different methodologies in comparison or proposals of use of different statistical techniques in algorithms’ comparison.In this paper, we focus our study on the use of statistical techniques in the analysis of evolutionary algorithms’ behaviour over optimization problems. A study about the required conditions for statistical analysis of the results is presented by using some models of evolutionary algorithms for real-coding optimization. This study is conducted in two ways: single-problem analysis and multiple-problem analysis. The results obtained state that a parametric statistical analysis could not be appropriate specially when we deal with multiple-problem results. In multiple-problem analysis, we propose the use of non-parametric statistical tests given that they are less restrictive than parametric ones and they can be used over small size samples of results. As a case study, we analyze the published results for the algorithms presented in the CEC’2005 Special Session on Real Parameter Optimization by using non-parametric test procedures.

A Web-Based Tool For Control Engineering Teaching In this article a new tool for control engineering teaching is presented. The tool was implemented using Java applets and is freely accessible through Web. It allows the analysis and simulation of linear control systems and was created to complement the theoretical lectures in basic control engineering courses. The article is not only centered in the description of the tool but also in the methodology to use it and its evaluation in an electrical engineering degree. Two practical problems are included in the manuscript to illustrate the use of the main functions implemented. The developed web-based tool can be accessed through the link http://www.controlweb.cyc.ull.es. (C) 2006 Wiley Periodicals, Inc.

Biologically-inspired soft exosuit. In this paper, we present the design and evaluation of a novel soft cable-driven exosuit that can apply forces to the body to assist walking. Unlike traditional exoskeletons which contain rigid framing elements, the soft exosuit is worn like clothing, yet can generate moments at the ankle and hip with magnitudes of 18% and 30% of those naturally generated by the body during walking, respectively. Our design uses geared motors to pull on Bowden cables connected to the suit near the ankle. The suit has the advantages over a traditional exoskeleton in that the wearer's joints are unconstrained by external rigid structures, and the worn part of the suit is extremely light, which minimizes the suit's unintentional interference with the body's natural biomechanics. However, a soft suit presents challenges related to actuation force transfer and control, since the body is compliant and cannot support large pressures comfortably. We discuss the design of the suit and actuation system, including principles by which soft suits can transfer force to the body effectively and the biological inspiration for the design. For a soft exosuit, an important design parameter is the combined effective stiffness of the suit and its interface to the wearer. We characterize the exosuit's effective stiffness, and present preliminary results from it generating assistive torques to a subject during walking. We envision such an exosuit having broad applicability for assisting healthy individuals as well as those with muscle weakness.

Robust Sparse Linear Discriminant Analysis Linear discriminant analysis (LDA) is a very popular supervised feature extraction method and has been extended to different variants. However, classical LDA has the following problems: 1) The obtained discriminant projection does not have good interpretability for features. 2) LDA is sensitive to noise. 3) LDA is sensitive to the selection of number of projection directions. In this paper, a novel feature extraction method called robust sparse linear discriminant analysis (RSLDA) is proposed to solve the above problems. Specifically, RSLDA adaptively selects the most discriminative features for discriminant analysis by introducing the l2;1 norm. An orthogonal matrix and a sparse matrix are also simultaneously introduced to guarantee that the extracted features can hold the main energy of the original data and enhance the robustness to noise, and thus RSLDA has the potential to perform better than other discriminant methods. Extensive experiments on six databases demonstrate that the proposed method achieves the competitive performance compared with other state-of-the-art feature extraction methods. Moreover, the proposed method is robust to the noisy data. IEEE

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

Beam Weights Optimization Based on Reinforcement Learning for 6G With the continuous evolution of wireless radio access network architecture, artificial intelligence (AI) plays an increasingly significant role in network optimization. Native-AI will realize fast and accurate optimization, real-time policy adjustment and trend prediction in future 6G systems. To cope with the increasing user density and complex wireless environment, beam weights optimization becomes one of the key factors to promote the communication capacity, coverage and anti-interference ability. In this paper, we build an intelligent network model based on the reinforcement learning (RL) algorithm and optimize the beam pattern in a real typical scenario that contains both roads and buildings. A field test is conducted to investigate the potential impacts of the intelligent algorithm on network optimization. The data used in the field test are real measurement reports from UEs, which imply the location information and the uneven distribution of UEs. The coverage performance and received signal quality are significantly improved after optimization. We also analyze the limitations of AI in the current 5G systems and discuss the future work on native-AI and RL.

The Sybil Attack Large-scale peer-to-peer systems facesecurity threats from faulty or hostile remotecomputing elements. To resist these threats, manysuch systems employ redundancy. However, if asingle faulty entity can present multiple identities,it can control a substantial fraction of the system,thereby undermining this redundancy. Oneapproach to preventing these &quot;Sybil attacks&quot; is tohave a trusted agency certify identities. Thispaper shows that, without a logically centralizedauthority, Sybil...

BLEU: a method for automatic evaluation of machine translation Human evaluations of machine translation are extensive but expensive. Human evaluations can take months to finish and involve human labor that can not be reused. We propose a method of automatic machine translation evaluation that is quick, inexpensive, and language-independent, that correlates highly with human evaluation, and that has little marginal cost per run. We present this method as an automated understudy to skilled human judges which substitutes for them when there is need for quick or frequent evaluations.

Computational thinking Summary form only given. My vision for the 21st century, Computational Thinking, will be a fundamental skill used by everyone in the world. To reading, writing, and arithmetic, we should add computational thinking to every child's analytical ability. Computational thinking involves solving problems, designing systems, and understanding human behavior by drawing on the concepts fundamental to computer science. Thinking like a computer scientist means more than being able to program a computer. It requires the ability to abstract and thus to think at multiple levels of abstraction. In this talk I will give many examples of computational thinking, argue that it has already influenced other disciplines, and promote the idea that teaching computational thinking can not only inspire future generations to enter the field of computer science but benefit people in all fields.

Fuzzy logic in control systems: fuzzy logic controller. I.

Switching between stabilizing controllers This paper deals with the problem of switching between several linear time-invariant (LTI) controllers—all of them capable of stabilizing a speci4c LTI process—in such a way that the stability of the closed-loop system is guaranteed for any switching sequence. We show that it is possible to 4nd realizations for any given family of controller transfer matrices so that the closed-loop system remains stable, no matter how we switch among the controller. The motivation for this problem is the control of complex systems where con8icting requirements make a single LTI controller unsuitable. ? 2002 Published by Elsevier Science Ltd.

Tabu Search - Part I

Bidirectional recurrent neural networks In the first part of this paper, a regular recurrent neural network (RNN) is extended to a bidirectional recurrent neural network (BRNN). The BRNN can be trained without the limitation of using input information just up to a preset future frame. This is accomplished by training it simultaneously in positive and negative time direction. Structure and training procedure of the proposed network are explained. In regression and classification experiments on artificial data, the proposed structure gives better results than other approaches. For real data, classification experiments for phonemes from the TIMIT database show the same tendency. In the second part of this paper, it is shown how the proposed bidirectional structure can be easily modified to allow efficient estimation of the conditional posterior probability of complete symbol sequences without making any explicit assumption about the shape of the distribution. For this part, experiments on real data are reported

An intensive survey of fair non-repudiation protocols With the phenomenal growth of the Internet and open networks in general, security services, such as non-repudiation, become crucial to many applications. Non-repudiation services must ensure that when Alice sends some information to Bob over a network, neither Alice nor Bob can deny having participated in a part or the whole of this communication. Therefore a fair non-repudiation protocol has to generate non-repudiation of origin evidences intended to Bob, and non-repudiation of receipt evidences destined to Alice. In this paper, we clearly define the properties a fair non-repudiation protocol must respect, and give a survey of the most important non-repudiation protocols without and with trusted third party (TTP). For the later ones we discuss the evolution of the TTP's involvement and, between others, describe the most recent protocol using a transparent TTP. We also discuss some ad-hoc problems related to the management of non-repudiation evidences.

Dynamic movement and positioning of embodied agents in multiparty conversations For embodied agents to engage in realistic multiparty conversation, they must stand in appropriate places with respect to other agents and the environment. When these factors change, such as an agent joining the conversation, the agents must dynamically move to a new location and/or orientation to accommodate. This paper presents an algorithm for simulating movement of agents based on observed human behavior using techniques developed for pedestrian movement in crowd simulations. We extend a previous group conversation simulation to include an agent motion algorithm. We examine several test cases and show how the simulation generates results that mirror real-life conversation settings.

An improved genetic algorithm with conditional genetic operators and its application to set-covering problem The genetic algorithm (GA) is a popular, biologically inspired optimization method. However, in the GA there is no rule of thumb to design the GA operators and select GA parameters. Instead, trial-and-error has to be applied. In this paper we present an improved genetic algorithm in which crossover and mutation are performed conditionally instead of probability. Because there are no crossover rate and mutation rate to be selected, the proposed improved GA can be more easily applied to a problem than the conventional genetic algorithms. The proposed improved genetic algorithm is applied to solve the set-covering problem. Experimental studies show that the improved GA produces better results over the conventional one and other methods.

Lane-level traffic estimations using microscopic traffic variables This paper proposes a novel inference method to estimate lane-level traffic flow, time occupancy and vehicle inter-arrival time on road segments where local information could not be measured and assessed directly. The main contributions of the proposed method are 1) the ability to perform lane-level estimations of traffic flow, time occupancy and vehicle inter-arrival time and 2) the ability to adapt to different traffic regimes by assessing only microscopic traffic variables. We propose a modified Kriging estimation model which explicitly takes into account both spatial and temporal variability. Performance evaluations are conducted using real-world data under different traffic regimes and it is shown that the proposed method outperforms a Kalman filter-based approach.

Convolutional Neural Network-Based Classification of Driver's Emotion during Aggressive and Smooth Driving Using Multi-Modal Camera Sensors. Because aggressive driving often causes large-scale loss of life and property, techniques for advance detection of adverse driver emotional states have become important for the prevention of aggressive driving behaviors. Previous studies have primarily focused on systems for detecting aggressive driver emotion via smart-phone accelerometers and gyro-sensors, or they focused on methods of detecting physiological signals using electroencephalography (EEG) or electrocardiogram (ECG) sensors. Because EEG and ECG sensors cause discomfort to drivers and can be detached from the driver's body, it becomes difficult to focus on bio-signals to determine their emotional state. Gyro-sensors and accelerometers depend on the performance of GPS receivers and cannot be used in areas where GPS signals are blocked. Moreover, if driving on a mountain road with many quick turns, a driver's emotional state can easily be misrecognized as that of an aggressive driver. To resolve these problems, we propose a convolutional neural network (CNN)-based method of detecting emotion to identify aggressive driving using input images of the driver's face, obtained using near-infrared (NIR) light and thermal camera sensors. In this research, we conducted an experiment using our own database, which provides a high classification accuracy for detecting driver emotion leading to either aggressive or smooth (i.e., relaxed) driving. Our proposed method demonstrates better performance than existing methods.

Ethical Considerations Of Applying Robots In Kindergarten Settings: Towards An Approach From A Macroperspective In child-robot interaction (cHRI) research, many studies pursue the goal to develop interactive systems that can be applied in everyday settings. For early education, increasingly, the setting of a kindergarten is targeted. However, when cHRI and research are brought into a kindergarten, a range of ethical and related procedural aspects have to be considered and dealt with. While ethical models elaborated within other human-robot interaction settings, e.g., assisted living contexts, can provide some important indicators for relevant issues, we argue that it is important to start developing a systematic approach to identify and tackle those ethical issues which rise with cHRI in kindergarten settings on a more global level and address the impact of the technology from a macroperspective beyond the effects on the individual. Based on our experience in conducting studies with children in general and pedagogical considerations on the role of the institution of kindergarten in specific, in this paper, we enfold some relevant aspects that have barely been addressed in an explicit way in current cHRI research. Four areas are analyzed and key ethical issues are identified in each area: (1) the institutional setting of a kindergarten, (2) children as a vulnerable group, (3) the caregivers' role, and (4) pedagogical concepts. With our considerations, we aim at (i) broadening the methodology of the current studies within the area of cHRI, (ii) revalidate it based on our comprehensive empirical experience with research in kindergarten settings, both laboratory and real-world contexts, and (iii) provide a framework for the development of a more systematic approach to address the ethical issues in cHRI research within kindergarten settings.

Stability Analysis and Estimation of Domain of Attraction for Positive Polynomial Fuzzy Systems With Input Saturation AbstractIn this paper, the stability and positivity of positive polynomial fuzzy model based (PPFMB) control system are investigated, in which the positive polynomial fuzzy model and positive polynomial fuzzy controller are allowed to have different premise membership functions from each other. These mismatched premise membership functions can increase the flexibility of controller design; however, it will lead to the conservative results when the stability is analyzed based on the Lyapunov stability theory. To relax the positivity/stability conditions, the improved Taylor-series-membership-functions-dependent (ITSMFD) method is introduced by introducing the sample points information of Taylor-series approximate membership functions, local error information and boundary information of substate space of premise variables into the stability/positivity conditions. Meanwhile, the ITSMFD method is extended to the PPFMB control system with input saturation to relax the estimation of domain of attraction. Finally, simulation examples are presented to verify the feasibility of this method.

Mobile cloud computing: A survey Despite increasing usage of mobile computing, exploiting its full potential is difficult due to its inherent problems such as resource scarcity, frequent disconnections, and mobility. Mobile cloud computing can address these problems by executing mobile applications on resource providers external to the mobile device. In this paper, we provide an extensive survey of mobile cloud computing research, while highlighting the specific concerns in mobile cloud computing. We present a taxonomy based on the key issues in this area, and discuss the different approaches taken to tackle these issues. We conclude the paper with a critical analysis of challenges that have not yet been fully met, and highlight directions for future work.

Harmony search algorithm for solving Sudoku Harmony search (HS) algorithm was applied to solving Sudoku puzzle. The HS is an evolutionary algorithm which mimics musicians' behaviors such as random play, memory-based play, and pitch-adjusted play when they perform improvisation. Sudoku puzzles in this study were formulated as an optimization problem with number-uniqueness penalties. HS could successfully solve the optimization problem after 285 function evaluations, taking 9 seconds. Also, sensitivity analysis of HS parameters was performed to obtain a better idea of algorithm parameter values.

DEC: dynamically evolving clustering and its application to structure identification of evolving fuzzy models. Identification of models from input-output data essentially requires estimation of appropriate cluster centers. In this paper, a new online evolving clustering approach for streaming data is proposed. Unlike other approaches that consider either the data density or distance from existing cluster centers, this approach uses cluster weight and distance before generating new clusters. To capture the dynamics of the data stream, the cluster weight is defined in both data and time space in such a way that it decays exponentially with time. It also applies concepts from computational geometry to determine the neighborhood information while forming clusters. A distinction is made between core and noncore clusters to effectively identify the real outliers. The approach efficiently estimates cluster centers upon which evolving Takagi-Sugeno models are developed. The experimental results with developed models show that the proposed approach attains results at par or better than existing approaches and significantly reduces the computational overhead, which makes it suitable for real-time applications.

An Easily Understandable Grey Wolf Optimizer and Its Application to Fuzzy Controller Tuning. This paper proposes an easily understandable Grey Wolf Optimizer (GWO) applied to the optimal tuning of the parameters of Takagi-Sugeno proportional-integral fuzzy controllers (T-S PI-FCs). GWO is employed for solving optimization problems focused on the minimization of discrete-time objective functions defined as the weighted sum of the absolute value of the control error and of the squared output sensitivity function, and the vector variable consists of the tuning parameters of the T-S PI-FCs. Since the sensitivity functions are introduced with respect to the parametric variations of the process, solving these optimization problems is important as it leads to fuzzy control systems with a reduced process parametric sensitivity obtained by a GWO-based fuzzy controller tuning approach. GWO algorithms applied with this regard are formulated in easily understandable terms for both vector and scalar operations, and discussions on stability, convergence, and parameter settings are offered. The controlled processes referred to in the course of this paper belong to a family of nonlinear servo systems, which are modeled by second order dynamics plus a saturation and dead zone static nonlinearity. Experimental results concerning the angular position control of a laboratory servo system are included for validating the proposed method.

Evolutionary Wall-Following Hexapod Robot Using Advanced Multiobjective Continuous Ant Colony Optimized Fuzzy Controller. This paper proposes an evolutionary wall-following hexapod robot, where a new multiobjective evolutionary fuzzy control approach is proposed to control both walking orientation and speed of a hexapod robot for a wall-following task. According to the measurements of four distance sensors, a fuzzy controller (FC) controls the walking speed of the robot by changing the common swing angles of its six ...

Fuzzy Logic in Dynamic Parameter Adaptation of Harmony Search Optimization for Benchmark Functions and Fuzzy Controllers. Nowadays the use of fuzzy logic has been increasing in popularity, and this is mainly due to the inference mechanism that allows simulating human reasoning in knowledge-based systems. The main contribution of this work is using the concepts of fuzzy logic in a method for dynamically adapting the main parameters of the harmony search algorithm during execution. Dynamical adaptation of parameters in metaheuristics has been shown to improve performance and accuracy in a wide range of applications. For this reason, we propose and approach for fuzzy adaptation of parameters in harmony search. Two case studies are considered for testing the proposed approach, the optimization of mathematical functions, which are unimodal, multimodal, hybrid, and composite functions and a control problem without noise and when noise is considered. A statistical comparison between the harmony search algorithm and the fuzzy harmony search algorithm is presented to verify the advantages of the proposed approach.

Finite-Time Input-to-State Stability and Applications to Finite-Time Control Design This paper extends the well-known concept, Sontag's input-to-state stability (ISS), to finite-time control problems. In other words, a new concept, finite-time input-to-state stability (FTISS), is proposed and then is applied to both the analysis of finite-time stability and the design of finite-time stabilizing feedback laws of control systems. With finite-time stability, nonsmoothness has to be considered, and serious technical challenges arise in the design of finite-time controllers and the stability analysis of the closed-loop system. It is found that FTISS plays an important role as the conventional ISS in the context of asymptotic stability analysis and smooth feedback stabilization. Moreover, a robust adaptive controller is proposed to handle nonlinear systems with parametric and dynamic uncertainties by virtue of FTISS and related arguments.

Adam: A Method for Stochastic Optimization. We introduce Adam, an algorithm for first-order gradient-based optimization of stochastic objective functions, based on adaptive estimates of lower-order moments. The method is straightforward to implement, is computationally efficient, has little memory requirements, is invariant to diagonal rescaling of the gradients, and is well suited for problems that are large in terms of data and/or parameters. The method is also appropriate for non-stationary objectives and problems with very noisy and/or sparse gradients. The hyper-parameters have intuitive interpretations and typically require little tuning. Some connections to related algorithms, on which Adam was inspired, are discussed. We also analyze the theoretical convergence properties of the algorithm and provide a regret bound on the convergence rate that is comparable to the best known results under the online convex optimization framework. Empirical results demonstrate that Adam works well in practice and compares favorably to other stochastic optimization methods. Finally, we discuss AdaMax, a variant of Adam based on the infinity norm.

Multiple Lyapunov functions and other analysis tools for switched and hybrid systems In this paper, we introduce some analysis tools for switched and hybrid systems. We first present work on stability analysis. We introduce multiple Lyapunov functions as a tool for analyzing Lyapunov stability and use iterated function systems (IFS) theory as a tool for Lagrange stability. We also discuss the case where the switched systems are indexed by an arbitrary compact set. Finally, we extend Bendixson's theorem to the case of Lipschitz continuous vector fields, allowing limit cycle analysis of a class of "continuous switched" systems.

Learning to Predict Driver Route and Destination Intent For many people, driving is a routine activity where people drive to the same destinations using the same routes on a regular basis. Many drivers, for example, will drive to and from work along a small set of routes, at about the same time every day of the working week. Similarly, although a person may shop on different days or at different times, they will often visit the same grocery store(s). In this paper, we present a novel approach to predicting driver intent that exploits the predictable nature of everyday driving. Our approach predicts a driver's intended route and destination through the use of a probabilistic model learned from observation of their driving habits. We show that by using a low-cost GPS sensor and a map database, it is possible to build a hidden Markov model (HMM) of the routes and destinations used by the driver. Furthermore, we show that this model can be used to make accurate predictions of the driver's destination and route through on-line observation of their GPS position during the trip. We present a thorough evaluation of our approach using a corpus of almost a month of real, everyday driving. Our results demonstrate the effectiveness of the approach, achieving approximately 98% accuracy in most cases. Such high performance suggests that the method can be harnessed for improved safety monitoring, route planning taking into account traffic density, and better trip duration prediction

Software-Defined Networking: A Comprehensive Survey The Internet has led to the creation of a digital society, where (almost) everything is connected and is accessible from anywhere. However, despite their widespread adoption, traditional IP networks are complex and very hard to manage. It is both difficult to configure the network according to predefined policies, and to reconfigure it to respond to faults, load, and changes. To make matters even more difficult, current networks are also vertically integrated: the control and data planes are bundled together. Software-defined networking (SDN) is an emerging paradigm that promises to change this state of affairs, by breaking vertical integration, separating the network's control logic from the underlying routers and switches, promoting (logical) centralization of network control, and introducing the ability to program the network. The separation of concerns, introduced between the definition of network policies, their implementation in switching hardware, and the forwarding of traffic, is key to the desired flexibility: by breaking the network control problem into tractable pieces, SDN makes it easier to create and introduce new abstractions in networking, simplifying network management and facilitating network evolution. In this paper, we present a comprehensive survey on SDN. We start by introducing the motivation for SDN, explain its main concepts and how it differs from traditional networking, its roots, and the standardization activities regarding this novel paradigm. Next, we present the key building blocks of an SDN infrastructure using a bottom-up, layered approach. We provide an in-depth analysis of the hardware infrastructure, southbound and northbound application programming interfaces (APIs), network virtualization layers, network operating systems (SDN controllers), network programming languages, and network applications. We also look at cross-layer problems such as debugging and troubleshooting. In an effort to anticipate the future evolution of this - ew paradigm, we discuss the main ongoing research efforts and challenges of SDN. In particular, we address the design of switches and control platforms - with a focus on aspects such as resiliency, scalability, performance, security, and dependability - as well as new opportunities for carrier transport networks and cloud providers. Last but not least, we analyze the position of SDN as a key enabler of a software-defined environment.

Deep Learning in Mobile and Wireless Networking: A Survey. The rapid uptake of mobile devices and the rising popularity of mobile applications and services pose unprecedented demands on mobile and wireless networking infrastructure. Upcoming 5G systems are evolving to support exploding mobile traffic volumes, real-time extraction of fine-grained analytics, and agile management of network resources, so as to maximize user experience. Fulfilling these tasks is challenging, as mobile environments are increasingly complex, heterogeneous, and evolving. One potential solution is to resort to advanced machine learning techniques, in order to help manage the rise in data volumes and algorithm-driven applications. The recent success of deep learning underpins new and powerful tools that tackle problems in this space. In this paper we bridge the gap between deep learning and mobile and wireless networking research, by presenting a comprehensive survey of the crossovers between the two areas. We first briefly introduce essential background and state-of-the-art in deep learning techniques with potential applications to networking. We then discuss several techniques and platforms that facilitate the efficient deployment of deep learning onto mobile systems. Subsequently, we provide an encyclopedic review of mobile and wireless networking research based on deep learning, which we categorize by different domains. Drawing from our experience, we discuss how to tailor deep learning to mobile environments. We complete this survey by pinpointing current challenges and open future directions for research.

Attitudes Towards Social Robots In Education: Enthusiast, Practical, Troubled, Sceptic, And Mindfully Positive While social robots bring new opportunities for education, they also come with moral challenges. Therefore, there is a need for moral guidelines for the responsible implementation of these robots. When developing such guidelines, it is important to include different stakeholder perspectives. Existing (qualitative) studies regarding these perspectives however mainly focus on single stakeholders. In this exploratory study, we examine and compare the attitudes of multiple stakeholders on the use of social robots in primary education, using a novel questionnaire that covers various aspects of moral issues mentioned in earlier studies. Furthermore, we also group the stakeholders based on similarities in attitudes and examine which socio-demographic characteristics influence these attitude types. Based on the results, we identify five distinct attitude profiles and show that the probability of belonging to a specific profile is affected by such characteristics as stakeholder type, age, education and income. Our results also indicate that social robots have the potential to be implemented in education in a morally responsible way that takes into account the attitudes of various stakeholders, although there are multiple moral issues that need to be addressed first. Finally, we present seven (practical) implications for a responsible application of social robots in education following from our results. These implications provide valuable insights into how social robots should be implemented.

Neural Congestion Prediction System For Trip Modelling In Heterogeneous Spatio-Temporal Patterns Until recently, urban cities have faced an increasing demand for an efficient system able to help drivers to discover the congested roads and avoid the long queues. In this paper, an Intelligent Traffic Congestion Prediction System (ITCPS) was developed to predict traffic congestion states in roads. The system embeds a Neural Network architecture able to handle the variation of traffic changes. It takes into account various traffic patterns in urban regions as well as highways during workdays and free-days. The developed system provides drivers with the fastest path and the estimated travel time to reach their destination. The performance of the developed system was tested using a big and real-world Global Positioning System (GPS) database gathered from vehicles circulating in Sfax city urban areas, Tunisia as well as the highways linking Sfax and other Tunisian cities. The results of congestion and travel time prediction provided by our system show promise when compared to other non-parametric techniques. Moreover, our model performs well even in cross-regions whose data were not used during training phase.

Knowledge harvesting in the big-data era The proliferation of knowledge-sharing communities such as Wikipedia and the progress in scalable information extraction from Web and text sources have enabled the automatic construction of very large knowledge bases. Endeavors of this kind include projects such as DBpedia, Freebase, KnowItAll, ReadTheWeb, and YAGO. These projects provide automatically constructed knowledge bases of facts about named entities, their semantic classes, and their mutual relationships. They contain millions of entities and hundreds of millions of facts about them. Such world knowledge in turn enables cognitive applications and knowledge-centric services like disambiguating natural-language text, semantic search for entities and relations in Web and enterprise data, and entity-oriented analytics over unstructured contents. Prominent examples of how knowledge bases can be harnessed include the Google Knowledge Graph and the IBM Watson question answering system. This tutorial presents state-of-the-art methods, recent advances, research opportunities, and open challenges along this avenue of knowledge harvesting and its applications. Particular emphasis will be on the twofold role of knowledge bases for big-data analytics: using scalable distributed algorithms for harvesting knowledge from Web and text sources, and leveraging entity-centric knowledge for deeper interpretation of and better intelligence with Big Data.

Reservoir computing approaches to recurrent neural network training Echo State Networks and Liquid State Machines introduced a new paradigm in artificial recurrent neural network (RNN) training, where an RNN (the reservoir) is generated randomly and only a readout is trained. The paradigm, becoming known as reservoir computing, greatly facilitated the practical application of RNNs and outperformed classical fully trained RNNs in many tasks. It has lately become a vivid research field with numerous extensions of the basic idea, including reservoir adaptation, thus broadening the initial paradigm to using different methods for training the reservoir and the readout. This review systematically surveys both current ways of generating/adapting the reservoirs and training different types of readouts. It offers a natural conceptual classification of the techniques, which transcends boundaries of the current “brand-names” of reservoir methods, and thus aims to help in unifying the field and providing the reader with a detailed “map” of it.

Comment on "On Discriminative vs. Generative Classifiers: A Comparison of Logistic Regression and Naive Bayes" Comparison of generative and discriminative classifiers is an ever-lasting topic. As an important contribution to this topic, based on their theoretical and empirical comparisons between the naïve Bayes classifier and linear logistic regression, Ng and Jordan (NIPS 841---848, 2001) claimed that there exist two distinct regimes of performance between the generative and discriminative classifiers with regard to the training-set size. In this paper, our empirical and simulation studies, as a complement of their work, however, suggest that the existence of the two distinct regimes may not be so reliable. In addition, for real world datasets, so far there is no theoretically correct, general criterion for choosing between the discriminative and the generative approaches to classification of an observation x into a class y; the choice depends on the relative confidence we have in the correctness of the specification of either p(y|x) or p(x, y) for the data. This can be to some extent a demonstration of why Efron (J Am Stat Assoc 70(352):892---898, 1975) and O'Neill (J Am Stat Assoc 75(369):154---160, 1980) prefer normal-based linear discriminant analysis (LDA) when no model mis-specification occurs but other empirical studies may prefer linear logistic regression instead. Furthermore, we suggest that pairing of either LDA assuming a common diagonal covariance matrix (LDA-驴) or the naïve Bayes classifier and linear logistic regression may not be perfect, and hence it may not be reliable for any claim that was derived from the comparison between LDA-驴 or the naïve Bayes classifier and linear logistic regression to be generalised to all generative and discriminative classifiers.

Dest-ResNet: A Deep Spatiotemporal Residual Network for Hotspot Traffic Speed Prediction. With the ever-increasing urbanization process, the traffic jam has become a common problem in the metropolises around the world, making the traffic speed prediction a crucial and fundamental task. This task is difficult due to the dynamic and intrinsic complexity of the traffic environment in urban cities, yet the emergence of crowd map query data sheds new light on it. In general, a burst of crowd map queries for the same destination in a short duration (called "hotspot'') could lead to traffic congestion. For example, queries of the Capital Gym burst on weekend evenings lead to traffic jams around the gym. However, unleashing the power of crowd map queries is challenging due to the innate spatiotemporal characteristics of the crowd queries. To bridge the gap, this paper firstly discovers hotspots underlying crowd map queries. These discovered hotspots address the spatiotemporal variations. Then Dest-ResNet (Deep spatiotemporal Residual Network) is proposed for hotspot traffic speed prediction. Dest-ResNet is a sequence learning framework that jointly deals with two sequences in different modalities, i.e., the traffic speed sequence and the query sequence. The main idea of Dest-ResNet is to learn to explain and amend the errors caused when the unimodal information is applied individually. In this way, Dest-ResNet addresses the temporal causal correlation between queries and the traffic speed. As a result, Dest-ResNet shows a 30% relative boost over the state-of-the-art methods on real-world datasets from Baidu Map.

Deep Autoencoder Neural Networks for Short-Term Traffic Congestion Prediction of Transportation Networks. Traffic congestion prediction is critical for implementing intelligent transportation systems for improving the efficiency and capacity of transportation networks. However, despite its importance, traffic congestion prediction is severely less investigated compared to traffic flow prediction, which is partially due to the severe lack of large-scale high-quality traffic congestion data and advanced algorithms. This paper proposes an accessible and general workflow to acquire large-scale traffic congestion data and to create traffic congestion datasets based on image analysis. With this workflow we create a dataset named Seattle Area Traffic Congestion Status (SATCS) based on traffic congestion map snapshots from a publicly available online traffic service provider Washington State Department of Transportation. We then propose a deep autoencoder-based neural network model with symmetrical layers for the encoder and the decoder to learn temporal correlations of a transportation network and predicting traffic congestion. Our experimental results on the SATCS dataset show that the proposed DCPN model can efficiently and effectively learn temporal relationships of congestion levels of the transportation network for traffic congestion forecasting. Our method outperforms two other state-of-the-art neural network models in prediction performance, generalization capability, and computation efficiency.

Short-Term Traffic Prediction Based on DeepCluster in Large-Scale Road Networks Short-term traffic prediction (STTP) is one of the most critical capabilities in Intelligent Transportation Systems (ITS), which can be used to support driving decisions, alleviate traffic congestion and improve transportation efficiency. However, STTP of large-scale road networks remains challenging due to the difficulties of effectively modeling the diverse traffic patterns by high-dimensional time series. Therefore, this paper proposes a framework that involves a deep clustering method for STTP in large-scale road networks. The deep clustering method is employed to supervise the representation learning in a visualized way from the large unlabeled dataset. More specifically, to fully exploit the traffic periodicity, the raw series is first divided into a number of sub-series for triplet generation. The convolutional neural networks (CNNs) with triplet loss are utilized to extract the features of shape by transforming the series into visual images. The shape-based representations are then used to cluster road segments into groups. Thereafter, a model sharing strategy is further proposed to build recurrent NNs-based predictions through group-based models (GBMs). GBM is built for a type of traffic patterns, instead of one road segment exclusively or all road segments uniformly. Our framework can not only significantly reduce the number of prediction models, but also improve their generalization by virtue of being trained on more diverse examples. Furthermore, the proposed framework over a selected road network in Beijing is evaluated. Experiment results show that the deep clustering method can effectively cluster the road segments and GBM can achieve comparable prediction accuracy against the IBM with less number of prediction models.

Discovering spatio-temporal causal interactions in traffic data streams The detection of outliers in spatio-temporal traffic data is an important research problem in the data mining and knowledge discovery community. However to the best of our knowledge, the discovery of relationships, especially causal interactions, among detected traffic outliers has not been investigated before. In this paper we propose algorithms which construct outlier causality trees based on temporal and spatial properties of detected outliers. Frequent substructures of these causality trees reveal not only recurring interactions among spatio-temporal outliers, but potential flaws in the design of existing traffic networks. The effectiveness and strength of our algorithms are validated by experiments on a very large volume of real taxi trajectories in an urban road network.

A new approach for dynamic fuzzy logic parameter tuning in Ant Colony Optimization and its application in fuzzy control of a mobile robot Central idea is to avoid or slow down full convergence through the dynamic variation of parameters.Performance of different ACO variants was observed to choose one as the basis to the proposed approach.Convergence fuzzy controller with the objective of maintaining diversity to avoid premature convergence was created. Ant Colony Optimization is a population-based meta-heuristic that exploits a form of past performance memory that is inspired by the foraging behavior of real ants. The behavior of the Ant Colony Optimization algorithm is highly dependent on the values defined for its parameters. Adaptation and parameter control are recurring themes in the field of bio-inspired optimization algorithms. The present paper explores a new fuzzy approach for diversity control in Ant Colony Optimization. The main idea is to avoid or slow down full convergence through the dynamic variation of a particular parameter. The performance of different variants of the Ant Colony Optimization algorithm is analyzed to choose one as the basis to the proposed approach. A convergence fuzzy logic controller with the objective of maintaining diversity at some level to avoid premature convergence is created. Encouraging results on several traveling salesman problem instances and its application to the design of fuzzy controllers, in particular the optimization of membership functions for a unicycle mobile robot trajectory control are presented with the proposed method.

Adaptive Navigation Support Adaptive navigation support is a specific group of technologies that support user navigation in hyperspace, by adapting to the goals, preferences and knowledge of the individual user. These technologies, originally developed in the field of adaptive hypermedia, are becoming increasingly important in several adaptive Web applications, ranging from Web-based adaptive hypermedia to adaptive virtual reality. This chapter provides a brief introduction to adaptive navigation support, reviews major adaptive navigation support technologies and mechanisms, and illustrates these with a range of examples.

Learning to Predict Driver Route and Destination Intent For many people, driving is a routine activity where people drive to the same destinations using the same routes on a regular basis. Many drivers, for example, will drive to and from work along a small set of routes, at about the same time every day of the working week. Similarly, although a person may shop on different days or at different times, they will often visit the same grocery store(s). In this paper, we present a novel approach to predicting driver intent that exploits the predictable nature of everyday driving. Our approach predicts a driver's intended route and destination through the use of a probabilistic model learned from observation of their driving habits. We show that by using a low-cost GPS sensor and a map database, it is possible to build a hidden Markov model (HMM) of the routes and destinations used by the driver. Furthermore, we show that this model can be used to make accurate predictions of the driver's destination and route through on-line observation of their GPS position during the trip. We present a thorough evaluation of our approach using a corpus of almost a month of real, everyday driving. Our results demonstrate the effectiveness of the approach, achieving approximately 98% accuracy in most cases. Such high performance suggests that the method can be harnessed for improved safety monitoring, route planning taking into account traffic density, and better trip duration prediction

A Minimal Set Of Coordinates For Describing Humanoid Shoulder Motion The kinematics of the anatomical shoulder are analysed and modelled as a parallel mechanism similar to a Stewart platform. A new method is proposed to describe the shoulder kinematics with minimal coordinates and solve the indeterminacy. The minimal coordinates are defined from bony landmarks and the scapulothoracic kinematic constraints. Independent from one another, they uniquely characterise the shoulder motion. A humanoid mechanism is then proposed with identical kinematic properties. It is then shown how minimal coordinates can be obtained for this mechanism and how the coordinates simplify both the motion-planning task and trajectory-tracking control. Lastly, the coordinates are also shown to have an application in the field of biomechanics where they can be used to model the scapulohumeral rhythm.

Massive MIMO Antenna Selection: Switching Architectures, Capacity Bounds, and Optimal Antenna Selection Algorithms. Antenna selection is a multiple-input multiple-output (MIMO) technology, which uses radio frequency (RF) switches to select a good subset of antennas. Antenna selection can alleviate the requirement on the number of RF transceivers, thus being attractive for massive MIMO systems. In massive MIMO antenna selection systems, RF switching architectures need to be carefully considered. In this paper, w...

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

A Certificateless Authenticated Key Agreement Protocol for Digital Rights Management System.

An Improved RSA Based User Authentication and Session Key Agreement Protocol Usable in TMIS. Recently, Giri et al.'s proposed a RSA cryptosystem based remote user authentication scheme for telecare medical information system and claimed that the protocol is secure against all the relevant security attacks. However, we have scrutinized the Giri et al.'s protocol and pointed out that the protocol is not secure against off-line password guessing attack, privileged insider attack and also suffers from anonymity problem. Moreover, the extension of password guessing attack leads to more security weaknesses. Therefore, this protocol needs improvement in terms of security before implementing in real-life application. To fix the mentioned security pitfalls, this paper proposes an improved scheme over Giri et al.'s scheme, which preserves user anonymity property. We have then simulated the proposed protocol using widely-accepted AVISPA tool which ensures that the protocol is SAFE under OFMC and CL-AtSe models, that means the same protocol is secure against active and passive attacks including replay and man-in-the-middle attacks. The informal cryptanalysis has been also presented, which confirmed that the proposed protocol provides well security protection on the relevant security attacks. The performance analysis section compares the proposed protocol with other existing protocols in terms of security and it has been observed that the protocol provides more security and achieves additional functionalities such as user anonymity and session key verification.

A more secure digital rights management authentication scheme based on smart card Digital rights management (DRM) system is a technology based mechanism to ensure only authorized access and legal distribution/consumption of the protected digital content. DRM system deals with the whole lifecycle of the digital content including production, management, distribution and consumption. DRM schemes are effective means for the transfer of digital content and safeguard the intellectual property. Recently, Yang et al. proposed a smart-card based DRM authentication scheme providing mutual authentication and session key establishment among all the participants of the DRM environment. We show that their scheme does not resist threats like smart card attack; fails to provide proper password update facility; and does not follow forward secrecy. To overcome these weaknesses, we propose an improvement of Yang et al.’s scheme. The security of our scheme remains intact even if the smart card of the user is lost. In our scheme, user’s smart card is capable of verifying the correctness of the inputted identity and password and hence contributes to achieve an efficient and user- friendly password update phase. In addition, the session keys established between the participating entities are highly secure by virtue of forward secrecy property. We conduct security analysis and comparison with related schemes to evaluate our improved scheme. During comparison, we also highlight the computational cost/time complexity at the user and the server side in terms of the execution time of various operations. The entire analysis shows that the design of the improved scheme is robust enough for the for DRM environment.

An improved E-DRM scheme for mobile environments. With the rapid development of information science and network technology, Internet has become an important platform for the dissemination of digital content, which can be easily copied and distributed through the Internet. Although convenience is increased, it causes significant damage to authors of digital content. Digital rights management system (DRM system) is an access control system that is designed to protect digital content and ensure illegal users from maliciously spreading digital content. Enterprise Digital Rights Management system (E-DRM system) is a DRM system that prevents unauthorized users from stealing the enterprise's confidential data. User authentication is the most important method to ensure digital rights management. In order to verify the validity of user, the biometrics-based authentication protocol is widely used due to the biological characteristics of each user are unique. By using biometric identification, it can ensure the correctness of user identity. In addition, due to the popularity of mobile device and Internet, user can access digital content and network information at anytime and anywhere. Recently, Mishra et al. proposed an anonymous and secure biometric-based enterprise digital rights management system for mobile environment. Although biometrics-based authentication is used to prevent users from being forged, the anonymity of users and the preservation of digital content are not ensured in their proposed system. Therefore, in this paper, we will propose a more efficient and secure biometric-based enterprise digital rights management system with user anonymity for mobile environments.

A Robust Anonymous Authentication Scheme using Biometrics for Digital Rights Management System As digital content transmission through the internet is convenient and quick, so the outspread of digital content is very high. However, along with this incredible speed and ease, current communication technologies and computers have also brought with them plenty of digital rights management complications. Digital Rights Management Systems are designed to limit the access to the utilization, alter...

Computational Efficient Authenticated Digital Content Distribution Frameworks for DRM Systems: Review and Outlook Advancement in digital technologies presents a user-friendly environment for the digital content distribution. However, it makes digital content prone to piracy issues. Digital rights management (DRM) systems aim to ensure the authorized content usage. As the digital content broadcasts through the public network, a secure and authorized content access mechanism is required. As digital media users ...

Probabilistic encryption A new probabilistic model of data encryption is introduced. For this model, under suitable complexity assumptions, it is proved that extracting any information about the cleartext from the cyphertext is hard on the average for an adversary with polynomially bounded computational resources. The proof holds for any message space with any probability distribution. The first implementation of this model is presented. The security of this implementation is proved under the interactability assumptin of deciding Quadratic Residuosity modulo composite numbers whose factorization is unknown.

A powerful and efficient algorithm for numerical function optimization: artificial bee colony (ABC) algorithm Swarm intelligence is a research branch that models the population of interacting agents or swarms that are able to self-organize. An ant colony, a flock of birds or an immune system is a typical example of a swarm system. Bees' swarming around their hive is another example of swarm intelligence. Artificial Bee Colony (ABC) Algorithm is an optimization algorithm based on the intelligent behaviour of honey bee swarm. In this work, ABC algorithm is used for optimizing multivariable functions and the results produced by ABC, Genetic Algorithm (GA), Particle Swarm Algorithm (PSO) and Particle Swarm Inspired Evolutionary Algorithm (PS-EA) have been compared. The results showed that ABC outperforms the other algorithms.

Toward Massive Machine Type Communications in Ultra-Dense Cellular IoT Networks: Current Issues and Machine Learning-Assisted Solutions The ever-increasing number of resource-constrained machine-type communication (MTC) devices is leading to the critical challenge of fulfilling diverse communication requirements in dynamic and ultra-dense wireless environments. Among different application scenarios that the upcoming 5G and beyond cellular networks are expected to support, such as enhanced mobile broadband (eMBB), massive machine type communications (mMTCs), and ultra-reliable and low latency communications (URLLCs), the mMTC brings the unique technical challenge of supporting a huge number of MTC devices in cellular networks, which is the main focus of this paper. The related challenges include quality of service (QoS) provisioning, handling highly dynamic and sporadic MTC traffic, huge signalling overhead, and radio access network (RAN) congestion. In this regard, this paper aims to identify and analyze the involved technical issues, to review recent advances, to highlight potential solutions and to propose new research directions. First, starting with an overview of mMTC features and QoS provisioning issues, we present the key enablers for mMTC in cellular networks. Along with the highlights on the inefficiency of the legacy random access (RA) procedure in the mMTC scenario, we then present the key features and channel access mechanisms in the emerging cellular IoT standards, namely, LTE-M and narrowband IoT (NB-IoT). Subsequently, we present a framework for the performance analysis of transmission scheduling with the QoS support along with the issues involved in short data packet transmission. Next, we provide a detailed overview of the existing and emerging solutions toward addressing RAN congestion problem, and then identify potential advantages, challenges, and use cases for the applications of emerging machine learning (ML) techniques in ultra-dense cellular networks. Out of several ML techniques, we focus on the application of low-complexity <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula> -learning approach in the mMTC scenario along with the recent advances toward enhancing its learning performance and convergence. Finally, we discuss some open research challenges and promising future research directions.

Priced Oblivious Transfer: How to Sell Digital Goods We consider the question of protecting the privacy of customers buying digital goods. More specifically, our goal is to allow a buyer to purchase digital goods from a vendor without letting the vendor learn what, and to the extent possible also when and how much, it is buying. We propose solutions which allow the buyer, after making an initial deposit, to engage in an unlimited number of priced oblivious-transfer protocols, satisfying the following requirements: As long as the buyer's balance contains sufficient funds, it will successfully retrieve the selected item and its balance will be debited by the item's price. However, the buyer should be unable to retrieve an item whose cost exceeds its remaining balance. The vendor should learn nothing except what must inevitably be learned, namely, the amount of interaction and the initial deposit amount (which imply upper bounds on the quantity and total price of all information obtained by the buyer). In particular, the vendor should be unable to learn what the buyer's current balance is or when it actually runs out of its funds. The technical tools we develop, in the process of solving this problem, seem to be of independent interest. In particular, we present the first one-round (two-pass) protocol for oblivious transfer that does not rely on the random oracle model (a very similar protocol was independently proposed by Naor and Pinkas [21]). This protocol is a special case of a more general "conditional disclosure" methodology, which extends a previous approach from [11] and adapts it to the 2-party setting.

Minimum acceleration criterion with constraints implies bang-bang control as an underlying principle for optimal trajectories of arm reaching movements. Rapid arm-reaching movements serve as an excellent test bed for any theory about trajectory formation. How are these movements planned? A minimum acceleration criterion has been examined in the past, and the solution obtained, based on the Euler-Poisson equation, failed to predict that the hand would begin and end the movement at rest (i.e., with zero acceleration). Therefore, this criterion was rejected in favor of the minimum jerk, which was proved to be successful in describing many features of human movements. This letter follows an alternative approach and solves the minimum acceleration problem with constraints using Pontryagin's minimum principle. We use the minimum principle to obtain minimum acceleration trajectories and use the jerk as a control signal. In order to find a solution that does not include nonphysiological impulse functions, constraints on the maximum and minimum jerk values are assumed. The analytical solution provides a three-phase piecewise constant jerk signal (bang-bang control) where the magnitude of the jerk and the two switching times depend on the magnitude of the maximum and minimum available jerk values. This result fits the observed trajectories of reaching movements and takes into account both the extrinsic coordinates and the muscle limitations in a single framework. The minimum acceleration with constraints principle is discussed as a unifying approach for many observations about the neural control of movements.

Wireless Networks with RF Energy Harvesting: A Contemporary Survey Radio frequency (RF) energy transfer and harvesting techniques have recently become alternative methods to power the next generation wireless networks. As this emerging technology enables proactive energy replenishment of wireless devices, it is advantageous in supporting applications with quality of service (QoS) requirements. In this paper, we present a comprehensive literature review on the research progresses in wireless networks with RF energy harvesting capability, referred to as RF energy harvesting networks (RF-EHNs). First, we present an overview of the RF-EHNs including system architecture, RF energy harvesting techniques and existing applications. Then, we present the background in circuit design as well as the state-of-the-art circuitry implementations, and review the communication protocols specially designed for RF-EHNs. We also explore various key design issues in the development of RFEHNs according to the network types, i.e., single-hop networks, multi-antenna networks, relay networks, and cognitive radio networks. Finally, we envision some open research directions.

Automated Detection of Obstructive Sleep Apnea Events from a Single-Lead Electrocardiogram Using a Convolutional Neural Network. In this study, we propose a method for the automated detection of obstructive sleep apnea (OSA) from a single-lead electrocardiogram (ECG) using a convolutional neural network (CNN). A CNN model was designed with six optimized convolution layers including activation, pooling, and dropout layers. One-dimensional (1D) convolution, rectified linear units (ReLU), and max pooling were applied to the convolution, activation, and pooling layers, respectively. For training and evaluation of the CNN model, a single-lead ECG dataset was collected from 82 subjects with OSA and was divided into training (including data from 63 patients with 34,281 events) and testing (including data from 19 patients with 8571 events) datasets. Using this CNN model, a precision of 0.99%, a recall of 0.99%, and an F-score of 0.99% were attained with the training dataset; these values were all 0.96% when the CNN was applied to the testing dataset. These results show that the proposed CNN model can be used to detect OSA accurately on the basis of a single-lead ECG. Ultimately, this CNN model may be used as a screening tool for those suspected to suffer from OSA.

A Muscle Synergy-Driven ANFIS Approach to Predict Continuous Knee Joint Movement Continuous motion prediction plays a significant role in realizing seamless control of robotic exoskeletons and orthoses. Explicitly modeling the relationship between coordinated muscle activations from surface electromyography (sEMG) and human limb movements provides a new path of sEMG-based human–machine interface. Instead of the numeric features from individual channels, we propose a muscle synergy-driven adaptive network-based fuzzy inference system (ANFIS) approach to predict continuous knee joint movements, in which muscle synergy reflects the motor control information to coordinate muscle activations for performing movements. Four human subjects participated in the experiment while walking at five types of speed: 2.0 km/h, 2.5 km/h, 3.0 km/h, 3.5 km/h, and 4.0 km/h. The study finds that the acquired muscle synergies associate the muscle activations with human joint movements in a low-dimensional space and have been further utilized for predicting knee joint angles. The proposed approach outperformed commonly used numeric features from individual sEMG channels with an average correlation coefficient of 0.92 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$ \pm $</tex-math></inline-formula> 0.05. Results suggest that the correlation between muscle activations and knee joint movements is captured by the muscle synergy-driven ANFIS model and can be utilized for the estimation of continuous joint angles.

A Novel NLOS Mitigation Algorithm for UWB Localization in Harsh Indoor Environments. Non-line-of-sight (NLOS) propagation of radio signals can significantly degrade the performance of ultra-wideband localization systems indoors, it is hence crucial to mitigate the NLOS effect to enhance the accuracy of positioning. The existing NLOS mitigation algorithms to improve localization accuracy are either by compensating range errors through NLOS identification and mitigation methods for ...

Survey of NLOS identification and error mitigation problems in UWB-based positioning algorithms for dense environments In this survey, the currently available ultra-wideband-based non-line-of-sight (NLOS) identification and error mitigation methods are presented. They are classified into several categories and their comparison is presented in two tables: one each for NLOS identification and error mitigation. NLOS identification methods are classified based on range estimates, channel statistics, and the actual maps of the building and environment. NLOS error mitigation methods are categorized based on direct path and statistics-based detection.

Reliable Classification of Vehicle Types Based on Cascade Classifier Ensembles Vehicle-type recognition based on images is a challenging task. This paper comparatively studied two feature extraction methods for image description, i.e., the Gabor wavelet transform and the Pyramid Histogram of Oriented Gradients (PHOG). The Gabor transform has been widely adopted to extract image features for various vision tasks. PHOG has the superiority in its description of more discriminating information. A highly reliable classification scheme was proposed by cascade classifier ensembles with reject option to accommodate the situations where no decision should be made if there exists adequate ambiguity. The first ensemble is heterogeneous, consisting of several classifiers, including $k$-nearest neighbors (kNNs), multiple-layer perceptrons (MLPs), support vector machines (SVMs), and random forest. The classification reliability is further enhanced by a second classifier ensemble, which is composed of a set of base MLPs coordinated by an ensemble metalearning method called rotation forest (RF). For both of the ensembles, rejection option is accomplished by relating the consensus degree from majority voting to a confidence measure and by abstaining to classify ambiguous samples if the consensus degree is lower than a threshold. The final class label is assigned by dual majority voting from the two ensembles. Experimental results using more than 600 images from a variety of 21 makes of cars and vans demonstrated the effectiveness of the proposed approach. The cascade ensembles produce consistently reliable results. With a moderate ensemble size of 25 in the second ensemble, the two-stage classification scheme offers 98.65% accuracy with a rejection rate of 2.5%, exhibiting promising potential for real-world applications.

A fusion strategy for reliable vehicle positioning utilizing RFID and in-vehicle sensors. RFID is introduced as a virtual sensor for vehicle positioning.LSSVM algorithm is proposed to obtain the distance between RFID tags and reader.In-vehicle sensors are employed to fuse with RFID to achieve vehicle positioning.An LSSVM-MM (multiple models) filter is proposed to realize the global fusion. In recent years, RFID has become a viable solution to provide object's location information. However, the RFID-based positioning algorithms in the literature have disadvantages such as low accuracy, low output frequency and the lack of speed or attitude information. To overcome these problems, this paper proposes a RFID/in-vehicle sensors fusion strategy for vehicle positioning in completely GPS-denied environments such as tunnels. The low-cost in-vehicle sensors including electronic compass and wheel speed sensors are introduced to be fused with RFID. The strategy adopts a two-step approach, i.e., the calculation of the distances between the RFID tags and the reader, and then the global fusion estimation of vehicle position. First, a Least Square Support Vector Machine (LSSVM) algorithm is developed to obtain the distances. Further, a novel LSSVM Multiple Model (LMM) algorithm is designed to fuse the data obtained from RFID and in-vehicle sensors. Contrarily to other multiple model algorithms, the LMM is more suitable for current driving conditions because the model probabilities can be calculated according to the operating state of the vehicle by using the LSSVM decision model. Finally, the proposed strategy is evaluated through experiments. The results validate the feasibility and effectiveness of the proposed strategy. This paper proposes a RFID/in-vehicle sensors fusion strategy for vehicle positioning in completely GPS-denied environments such as tunnels. The low-cost in-vehicle sensors including electronic compass and wheel speed sensors are introduced to be fused with RFID. The strategy adopts a two-step approach, i.e., the calculation of the distances between the RFID tags and the reader, and then the global fusion estimation of vehicle position. First, a least square support vector machine (LSSVM) algorithm is developed to obtain the distance. Further, a novel LSSVM multiple model (LMM) algorithm is designed to fuse the data obtained from RFID and in-vehicle sensors. Contrarily to other multiple models algorithms, LMM is more suitable for current driving conditions because the model probabilities can be calculated according to the operating state of the vehicle by using the LSSVM decision model. Finally, the proposed strategy is evaluated through experiments. The results validate the feasibility and effectiveness of the proposed strategy.Display Omitted

Novel EKF-Based Vision/Inertial System Integration for Improved Navigation. With advances in computing power, stereo vision has become an essential part of navigation applications. However, there may be instances wherein insufficient image data precludes the estimation of navigation parameters. Earlier, a novel vision-based velocity estimation method was developed by the authors, which suffered from the aforementioned drawback. In this paper, the vision-based navigation m...

Relative Position Estimation Between Two UWB Devices With IMUs For a team of robots to work collaboratively, it is crucial that each robot have the ability to determine the position of their neighbors, relative to themselves, in order to execute tasks autonomously. This letter presents an algorithm for determining the three-dimensional relative position between two mobile robots, each using nothing more than a single ultra-wideband transceiver, an acceleromet...

Footprints: history-rich tools for information foraging Inspired by Hill and Hollans original work [7], we have beendeveloping a theory of interaction history and building tools toapply this theory to navigation in a complex information space. Wehave built a series of tools - map, paths, annota- tions andsignposts - based on a physical-world navigation metaphor. Thesetools have been in use for over a year. Our user study involved acontrolled browse task and showed that users were able to get thesame amount of work done with significantly less effort.

Very Deep Convolutional Networks for Large-Scale Image Recognition. In this work we investigate the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting. Our main contribution is a thorough evaluation of networks of increasing depth using an architecture with very small (3x3) convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers. These findings were the basis of our ImageNet Challenge 2014 submission, where our team secured the first and the second places in the localisation and classification tracks respectively. We also show that our representations generalise well to other datasets, where they achieve state-of-the-art results. We have made our two best-performing ConvNet models publicly available to facilitate further research on the use of deep visual representations in computer vision.

Chimp optimization algorithm. •A novel optimizer called Chimp Optimization Algorithm (ChOA) is proposed.•ChOA is inspired by individual intelligence and sexual motivation of chimps.•ChOA alleviates the problems of slow convergence rate and trapping in local optima.•The four main steps of Chimp hunting are implemented.

Space-time modeling of traffic flow. This paper discusses the application of space-time autoregressive integrated moving average (STARIMA) methodology for representing traffic flow patterns. Traffic flow data are in the form of spatial time series and are collected at specific locations at constant intervals of time. Important spatial characteristics of the space-time process are incorporated in the STARIMA model through the use of weighting matrices estimated on the basis of the distances among the various locations where data are collected. These matrices distinguish the space-time approach from the vector autoregressive moving average (VARMA) methodology and enable the model builders to control the number of the parameters that have to be estimated. The proposed models can be used for short-term forecasting of space-time stationary traffic-flow processes and for assessing the impact of traffic-flow changes on other parts of the network. The three-stage iterative space-time model building procedure is illustrated using 7.5min average traffic flow data for a set of 25 loop-detectors located at roads that direct to the centre of the city of Athens, Greece. Data for two months with different traffic-flow characteristics are modelled in order to determine the stability of the parameter estimation.

A novel full structure optimization algorithm for radial basis probabilistic neural networks. In this paper, a novel full structure optimization algorithm for radial basis probabilistic neural networks (RBPNN) is proposed. Firstly, a minimum volume covering hyperspheres (MVCH) algorithm is proposed to heuristically select the initial hidden layer centers of the RBPNN, and then the recursive orthogonal least square (ROLS) algorithm combined with the particle swarm optimization (PSO) algorithm is adopted to further optimize the initial structure of the RBPNN. Finally, the effectiveness and efficiency of our proposed algorithm are evaluated through a plant species identification task involving 50 plant species.

Understanding Taxi Service Strategies From Taxi GPS Traces Taxi service strategies, as the crowd intelligence of massive taxi drivers, are hidden in their historical time-stamped GPS traces. Mining GPS traces to understand the service strategies of skilled taxi drivers can benefit the drivers themselves, passengers, and city planners in a number of ways. This paper intends to uncover the efficient and inefficient taxi service strategies based on a large-scale GPS historical database of approximately 7600 taxis over one year in a city in China. First, we separate the GPS traces of individual taxi drivers and link them with the revenue generated. Second, we investigate the taxi service strategies from three perspectives, namely, passenger-searching strategies, passenger-delivery strategies, and service-region preference. Finally, we represent the taxi service strategies with a feature matrix and evaluate the correlation between service strategies and revenue, informing which strategies are efficient or inefficient. We predict the revenue of taxi drivers based on their strategies and achieve a prediction residual as less as 2.35 RMB/h,1 which demonstrates that the extracted taxi service strategies with our proposed approach well characterize the driving behavior and performance of taxi drivers.

Finite-Time Adaptive Fuzzy Tracking Control Design for Nonlinear Systems. This paper addresses the finite-time tracking problem of nonlinear pure-feedback systems. Unlike the literature on traditional finite-time stabilization, in this paper the nonlinear system functions, including the bounding functions, are all totally unknown. Fuzzy logic systems are used to model those unknown functions. To present a finite-time control strategy, a criterion of semiglobal practical...

Myoelectric or Force Control? A Comparative Study on a Soft Arm Exosuit The intention-detection strategy used to drive an exosuit is fundamental to evaluate the effectiveness and acceptability of the device. Yet, current literature on wearable soft robotics lacks evidence on the comparative performance of different control approaches for online intention-detection. In the present work, we compare two different and complementary controllers on a wearable robotic suit, previously formulated and tested by our group; a model-based myoelectric control ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">myoprocessor</i> ), which estimates the joint torque from the activation of target muscles, and a force control that estimates human torques using an inverse dynamics model ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">dynamic arm</i> ). We test them on a cohort of healthy participants performing tasks replicating functional activities of daily living involving a wide range of dynamic movements. Our results suggest that both controllers are robust and effective in detecting human–motor interaction, and show comparable performance for augmenting muscular activity. In particular, the biceps brachii activity was reduced by up to 74% under the assistance of the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">dynamic arm</i> and up to 47% under the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">myoprocessor</i> , compared to a no-suit condition. However, the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">myoprocessor</i> outperformed the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">dynamic arm</i> in promptness and assistance during movements that involve high dynamics. The exosuit work normalized with respect to the overall work was <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$68.84 \pm 3.81\%$</tex-math></inline-formula> when it was ran by the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">myoprocessor</i> , compared to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$45.29 \pm 7.71\%$</tex-math></inline-formula> during the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">dynamic arm</i> condition. The reliability and accuracy of motor intention detection strategies in wearable device is paramount for both the efficacy and acceptability of this technology. In this article, we offer a detailed analysis of the two most widely used control approaches, trying to highlight their intrinsic structural differences and to discuss their different and complementary performance.

A space-time delay neural network model for travel time prediction. Research on space-time modelling and forecasting has focused on integrating space-time autocorrelation into statistical models to increase the accuracy of forecasting. These models include space-time autoregressive integrated moving average (STARIMA) and its various extensions. However, they are inadequate for the cases when the correlation between data is dynamic and heterogeneous, such as traffic network data. The aim of the paper is to integrate spatial and temporal autocorrelations of road traffic network by developing a novel space-time delay neural network (STDNN) model that capture the autocorrelation locally and dynamically. Validation of the space-time delay neural network is carried out using real data from London road traffic network with 22 links by comparing benchmark models such as Naïve, ARIMA, and STARIMA models. Study results show that STDNN outperforms the Naïve, ARIMA, and STARIMA models in prediction accuracy and has considerable advantages in travel time prediction.

Analysis of feature selection stability on high dimension and small sample data Feature selection is an important step when building a classifier on high dimensional data. As the number of observations is small, the feature selection tends to be unstable. It is common that two feature subsets, obtained from different datasets but dealing with the same classification problem, do not overlap significantly. Although it is a crucial problem, few works have been done on the selection stability. The behavior of feature selection is analyzed in various conditions, not exclusively but with a focus on t -score based feature selection approaches and small sample data. The analysis is in three steps: the first one is theoretical using a simple mathematical model; the second one is empirical and based on artificial data; and the last one is based on real data. These three analyses lead to the same results and give a better understanding of the feature selection problem in high dimension data.

Discovering spatio-temporal dependencies based on time-lag in intelligent transportation data. Learning spatio-temporal dependency structure is meaningful to characterize causal or statistical relationships. In many real-world applications, dependency structure is often characterized by time-lag between variables. For example, traffic system and climate, time lag is a key feature of hidden temporal dependencies, and plays an essential role in interpreting the cause of discovered temporal dependencies. However, traditional dependencies learning algorithms only use the same time stamp data of variables. In this paper, we propose a method for mining dependencies by considering the time lag. The proposed approach is based on a decomposition of the coefficients into products of two-level hierarchical coefficients, where one represents feature-level and the other represents time-level. Specially, we capture the prior information of time lag in intelligent transportation data. We construct a probabilistic formulation by applying some probabilistic priors to these hierarchical coefficients, and devise an expectation-maximization (EM) algorithm to learn the model parameters. We evaluate our model on both synthetic and real-world highway traffic datasets. Experimental results show the effectiveness of our method.

Collective feature selection to identify crucial epistatic variants. In this study, we were able to show that selecting variables using a collective feature selection approach could help in selecting true positive epistatic variables more frequently than applying any single method for feature selection via simulation studies. We were able to demonstrate the effectiveness of collective feature selection along with a comparison of many methods in our simulation analysis. We also applied our method to identify non-linear networks associated with obesity.

Short-Term Traffic Forecasting by Mining the Non-Stationarity of Spatiotemporal Patterns Short-term traffic forecasting is important for the development of an intelligent traffic management system. Critical to the performance of the traffic prediction model utilized in such a system is accurate representation of the spatiotemporal traffic characteristics. This can be achieved by integrating spatiotemporal traffic information or the dynamic traffic characteristics in the modeling proce...

Subway Passenger Flow Prediction for Special Events Using Smart Card Data In order to reduce passenger delays and prevent severe overcrowding in the subway system, it is necessary to accurately predict the short-term passenger flow during special events. However, few studies have been conducted to predict the subway passenger flow under these conditions. Traditional methods, such as the autoregressive integrated moving average (ARIMA) model, were commonly used to analyze regular traffic demands. These methods usually neglected the volatility (heteroscedasticity) in passenger flow influenced by unexpected external factors. This paper, therefore, proposed a generic framework to analyze short-term passenger flow, considering the dynamic volatility and nonlinearity of passenger flow during special events. Four different generalized autoregressive conditional heteroscedasticity models, along with the ARIMA model, were used to model the mean and volatility of passenger flow based on the transit smart card data from two stations near the Olympic Sports Center, Nanjing, China. Multiple statistical methods were applied to evaluate the performance of the hybrid models. The results indicate that the volatility of passenger flow had significant nonlinear and asymmetric features during special events. The proposed framework could effectively capture the mean and volatility of passenger flow, and outperform the traditional methods in terms of accuracy and reliability. Overall, this paper can help transit agencies to better understand the deterministic and stochastic changes of the passenger flow, and implement precautionary countermeasures for large crowds in a subway station before and after special events.

Completely derandomized self-adaptation in evolution strategies. This paper puts forward two useful methods for self-adaptation of the mutation distribution - the concepts of derandomization and cumulation. Principle shortcomings of the concept of mutative strategy parameter control and two levels of derandomization are reviewed. Basic demands on the self-adaptation of arbitrary (normal) mutation distributions are developed. Applying arbitrary, normal mutation distributions is equivalent to applying a general, linear problem encoding. The underlying objective of mutative strategy parameter control is roughly to favor previously selected mutation steps in the future. If this objective is pursued rigorously, a completely derandomized self-adaptation scheme results, which adapts arbitrary normal mutation distributions. This scheme, called covariance matrix adaptation (CMA), meets the previously stated demands. It can still be considerably improved by cumulation - utilizing an evolution path rather than single search steps. Simulations on various test functions reveal local and global search properties of the evolution strategy with and without covariance matrix adaptation. Their performances are comparable only on perfectly scaled functions. On badly scaled, non-separable functions usually a speed up factor of several orders of magnitude is observed. On moderately mis-scaled functions a speed up factor of three to ten can be expected.

An online mechanism for multi-unit demand and its application to plug-in hybrid electric vehicle charging We develop an online mechanism for the allocation of an expiring resource to a dynamic agent population. Each agent has a non-increasing marginal valuation function for the resource, and an upper limit on the number of units that can be allocated in any period. We propose two versions on a truthful allocation mechanism. Each modifies the decisions of a greedy online assignment algorithm by sometimes cancelling an allocation of resources. One version makes this modification immediately upon an allocation decision while a second waits until the point at which an agent departs the market. Adopting a prior-free framework, we show that the second approach has better worst-case allocative efficiency and is more scalable. On the other hand, the first approach (with immediate cancellation) may be easier in practice because it does not need to reclaim units previously allocated. We consider an application to recharging plug-in hybrid electric vehicles (PHEVs). Using data from a real-world trial of PHEVs in the UK, we demonstrate higher system performance than a fixed price system, performance comparable with a standard, but non-truthful scheduling heuristic, and the ability to support 50% more vehicles at the same fuel cost than a simple randomized policy.

Blockchain Meets IoT: An Architecture for Scalable Access Management in IoT. The Internet of Things (IoT) is stepping out of its infancy into full maturity and establishing itself as a part of the future Internet. One of the technical challenges of having billions of devices deployed worldwide is the ability to manage them. Although access management technologies exist in IoT, they are based on centralized models which introduce a new variety of technical limitations to ma...

The contourlet transform: an efficient directional multiresolution image representation. The limitations of commonly used separable extensions of one-dimensional transforms, such as the Fourier and wavelet transforms, in capturing the geometry of image edges are well known. In this paper, we pursue a "true" two-dimensional transform that can capture the intrinsic geometrical structure that is key in visual information. The main challenge in exploring geometry in images comes from the discrete nature of the data. Thus, unlike other approaches, such as curvelets, that first develop a transform in the continuous domain and then discretize for sampled data, our approach starts with a discrete-domain construction and then studies its convergence to an expansion in the continuous domain. Specifically, we construct a discrete-domain multiresolution and multidirection expansion using nonseparable filter banks, in much the same way that wavelets were derived from filter banks. This construction results in a flexible multiresolution, local, and directional image expansion using contour segments, and, thus, it is named the contourlet transform. The discrete contourlet transform has a fast iterated filter bank algorithm that requires an order N operations for N-pixel images. Furthermore, we establish a precise link between the developed filter bank and the associated continuous-domain contourlet expansion via a directional multiresolution analysis framework. We show that with parabolic scaling and sufficient directional vanishing moments, contourlets achieve the optimal approximation rate for piecewise smooth functions with discontinuities along twice continuously differentiable curves. Finally, we show some numerical experiments demonstrating the potential of contourlets in several image processing applications. Index Terms-Contourlets, contours, filter banks, geometric image processing, multidirection, multiresolution, sparse representation, wavelets.

A novel full structure optimization algorithm for radial basis probabilistic neural networks. In this paper, a novel full structure optimization algorithm for radial basis probabilistic neural networks (RBPNN) is proposed. Firstly, a minimum volume covering hyperspheres (MVCH) algorithm is proposed to heuristically select the initial hidden layer centers of the RBPNN, and then the recursive orthogonal least square (ROLS) algorithm combined with the particle swarm optimization (PSO) algorithm is adopted to further optimize the initial structure of the RBPNN. Finally, the effectiveness and efficiency of our proposed algorithm are evaluated through a plant species identification task involving 50 plant species.

Online Prediction of Driver Distraction Based on Brain Activity Patterns This paper presents a new computational framework for early detection of driver distractions (map viewing) using brain activity measured by electroencephalographic (EEG) signals. Compared with most studies in the literature, which are mainly focused on the classification of distracted and nondistracted periods, this study proposes a new framework to prospectively predict the start and end of a distraction period, defined by map viewing. The proposed prediction algorithm was tested on a data set of continuous EEG signals recorded from 24 subjects. During the EEG recordings, the subjects were asked to drive from an initial position to a destination using a city map in a simulated driving environment. The overall accuracy values for the prediction of the start and the end of map viewing were 81% and 70%, respectively. The experimental results demonstrated that the proposed algorithm can predict the start and end of map viewing with relatively high accuracy and can be generalized to individual subjects. The outcome of this study has a high potential to improve the design of future intelligent navigation systems. Prediction of the start of map viewing can be used to provide route information based on a driver's needs and consequently avoid map-viewing activities. Prediction of the end of map viewing can be used to provide warnings for potential long map-viewing durations. Further development of the proposed framework and its applications in driver-distraction predictions are also discussed.

A blind medical image watermarking: DWT-SVD based robust and secure approach for telemedicine applications. In this paper, a blind image watermarking scheme based on discrete wavelet transform (DWT) and singular value decomposition (SVD) is proposed. In this scheme, DWT is applied on ROI (region of interest) of the medical image to get different frequency subbands of its wavelet decomposition. On the low frequency subband LL of the ROI, block-SVD is applied to get different singular matrices. A pair of elements with similar values is identified from the left singular value matrix of these selected blocks. The values of these pairs are modified using certain threshold to embed a bit of watermark content. Appropriate threshold is chosen to achieve the imperceptibility and robustness of medical image and watermark contents respectively. For authentication and identification of original medical image, one watermark image (logo) and other text watermark have been used. The watermark image provides authentication whereas the text data represents electronic patient record (EPR) for identification. At receiving end, blind recovery of both watermark contents is performed by a similar comparison scheme used during the embedding process. The proposed algorithm is applied on various groups of medical images like X-ray, CT scan and mammography. This scheme offers better visibility of watermarked image and recovery of watermark content due to DWT-SVD combination. Moreover, use of Hamming error correcting code (ECC) on EPR text bits reduces the BER and thus provides better recovery of EPR. The performance of proposed algorithm with EPR data coding by Hamming code is compared with the BCH error correcting code and it is found that later one perform better. A result analysis shows that imperceptibility of watermarked image is better as PSNR is above 43 dB and WPSNR is above 52 dB for all set of images. In addition, robustness of the scheme is better than existing scheme for similar set of medical images in terms of normalized correlation coefficient (NCC) and bit-error-rate (BER). An analysis is also carried out to verify the performance of the proposed scheme for different size of watermark contents (image and EPR data). It is observed from analysis that the proposed scheme is also appropriate for watermarking of color image. Using proposed scheme, watermark contents are extracted successfully under various noise attacks like JPEG compression, filtering, Gaussian noise, Salt and pepper noise, cropping, filtering and rotation. Performance comparison of proposed scheme with existing schemes shows proposed scheme has better robustness against different types of attacks. Moreover, the proposed scheme is also robust under set of benchmark attacks known as checkmark attacks.

Ethical Considerations Of Applying Robots In Kindergarten Settings: Towards An Approach From A Macroperspective In child-robot interaction (cHRI) research, many studies pursue the goal to develop interactive systems that can be applied in everyday settings. For early education, increasingly, the setting of a kindergarten is targeted. However, when cHRI and research are brought into a kindergarten, a range of ethical and related procedural aspects have to be considered and dealt with. While ethical models elaborated within other human-robot interaction settings, e.g., assisted living contexts, can provide some important indicators for relevant issues, we argue that it is important to start developing a systematic approach to identify and tackle those ethical issues which rise with cHRI in kindergarten settings on a more global level and address the impact of the technology from a macroperspective beyond the effects on the individual. Based on our experience in conducting studies with children in general and pedagogical considerations on the role of the institution of kindergarten in specific, in this paper, we enfold some relevant aspects that have barely been addressed in an explicit way in current cHRI research. Four areas are analyzed and key ethical issues are identified in each area: (1) the institutional setting of a kindergarten, (2) children as a vulnerable group, (3) the caregivers' role, and (4) pedagogical concepts. With our considerations, we aim at (i) broadening the methodology of the current studies within the area of cHRI, (ii) revalidate it based on our comprehensive empirical experience with research in kindergarten settings, both laboratory and real-world contexts, and (iii) provide a framework for the development of a more systematic approach to address the ethical issues in cHRI research within kindergarten settings.

Gender Bias in Coreference Resolution: Evaluation and Debiasing Methods. We introduce a new benchmark, WinoBias, for coreference resolution focused on gender bias. Our corpus contains Winograd-schema style sentences with entities corresponding to people referred by their occupation (e.g. the nurse, the doctor, the carpenter). We demonstrate that a rule-based, a feature-rich, and a neural coreference system all link gendered pronouns to pro-stereotypical entities with higher accuracy than anti-stereotypical entities, by an average difference of 21.1 in F1 score. Finally, we demonstrate a data-augmentation approach that, in combination with existing word-embedding debiasing techniques, removes the bias demonstrated by these systems in WinoBias without significantly affecting their performance on existing coreference benchmark datasets. Our dataset and code are available at this http URL

NLTK: the natural language toolkit The Natural Language Toolkit is a suite of program modules, data sets, tutorials and exercises, covering symbolic and statistical natural language processing. NLTK is written in Python and distributed under the GPL open source license. Over the past three years, NLTK has become popular in teaching and research. We describe the toolkit and report on its current state of development.

WP:clubhouse?: an exploration of Wikipedia's gender imbalance Wikipedia has rapidly become an invaluable destination for millions of information-seeking users. However, media reports suggest an important challenge: only a small fraction of Wikipedia's legion of volunteer editors are female. In the current work, we present a scientific exploration of the gender imbalance in the English Wikipedia's population of editors. We look at the nature of the imbalance itself, its effects on the quality of the encyclopedia, and several conflict-related factors that may be contributing to the gender gap. Our findings confirm the presence of a large gender gap among editors and a corresponding gender-oriented disparity in the content of Wikipedia's articles. Further, we find evidence hinting at a culture that may be resistant to female participation.

Understanding Back-Translation at Scale. An effective method to improve neural machine translation with monolingual data is to augment the parallel training corpus with back-translations of target language sentences. This work broadens the understanding of back-translation and investigates a number of methods to generate synthetic source sentences. We find that in all but resource poor settings back-translations obtained via sampling or noised beam outputs are most effective. Our analysis shows that sampling or noisy synthetic data gives a much stronger training signal than data generated by beam or greedy search. We also compare how synthetic data compares to genuine bitext and study various domain effects. Finally, we scale to hundreds of millions of monolingual sentences and achieve a new state of the art of 35 BLEU on the WMTu002714 English-German test set.

Language (Technology) is Power: A Critical Survey of "Bias" in NLP We survey 146 papers analyzing "bias" in NLP systems, finding that their motivations are often vague, inconsistent, and lacking in normative reasoning, despite the fact that analyzing "bias" is an inherently normative process. We further find that these papers' proposed quantitative techniques for measuring or mitigating "bias" are poorly matched to their motivations and do not engage with the relevant literature outside of NLP. Based on these findings, we describe the beginnings of a path forward by proposing three recommendations that should guide work analyzing "bias" in NLP systems. These recommendations rest on a greater recognition of the relationships between language and social hierarchies, encouraging researchers and practitioners to articulate their conceptualizations of "bias"---i.e., what kinds of system behaviors are harmful, in what ways, to whom, and why, as well as the normative reasoning underlying these statements---and to center work around the lived experiences of members of communities affected by NLP systems, while interrogating and reimagining the power relations between technologists and such communities.

Gender Bias in Contextualized Word Embeddings. In this paper, we quantify, analyze and mitigate gender bias exhibited in ELMou0027s contextualized word vectors. First, we conduct several intrinsic analyses and find that (1) training data for ELMo contains significantly more male than female entities, (2) the trained ELMo embeddings systematically encode gender information and (3) ELMo unequally encodes gender information about male and female entities. Then, we show that a state-of-the-art coreference system that depends on ELMo inherits its bias and demonstrates significant bias on the WinoBias probing corpus. Finally, we explore two methods to mitigate such gender bias and show that the bias demonstrated on WinoBias can be eliminated.

BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. We introduce a new language representation model called BERT, which stands for Bidirectional Encoder Representations from Transformers. Unlike recent language representation models, is designed to pre-train deep bidirectional representations by jointly conditioning on both left and right context in all layers. As a result, the pre-trained representations can be fine-tuned with just one additional output layer to create state-of-the-art models for a wide range of tasks, such as question answering and language inference, without substantial task-specific architecture modifications. BERT is conceptually simple and empirically powerful. It obtains new state-of-the-art results on eleven natural language processing tasks, including pushing the GLUE benchmark to 80.4% (7.6% absolute improvement), MultiNLI accuracy to 86.7 (5.6% absolute improvement) and the SQuAD v1.1 question answering Test F1 to 93.2 (1.5% absolute improvement), outperforming human performance by 2.0%.

Deep learning Deep learning allows computational models that are composed of multiple processing layers to learn representations of data with multiple levels of abstraction. These methods have dramatically improved the state-of-the-art in speech recognition, visual object recognition, object detection and many other domains such as drug discovery and genomics. Deep learning discovers intricate structure in large data sets by using the backpropagation algorithm to indicate how a machine should change its internal parameters that are used to compute the representation in each layer from the representation in the previous layer. Deep convolutional nets have brought about breakthroughs in processing images, video, speech and audio, whereas recurrent nets have shone light on sequential data such as text and speech. Machine-learning technology powers many aspects of modern society: from web searches to content filtering on social networks to recommendations on e-commerce websites, and it is increasingly present in consumer products such as cameras and smartphones. Machine-learning systems are used to identify objects in images, transcribe speech into text, match news items, posts or products with users' interests, and select relevant results of search. Increasingly, these applications make use of a class of techniques called deep learning. Conventional machine-learning techniques were limited in their ability to process natural data in their raw form. For decades, constructing a pattern-recognition or machine-learning system required careful engineering and considerable domain expertise to design a feature extractor that transformed the raw data (such as the pixel values of an image) into a suitable internal representation or feature vector from which the learning subsystem, often a classifier, could detect or classify patterns in the input. Representation learning is a set of methods that allows a machine to be fed with raw data and to automatically discover the representations needed for detection or classification. Deep-learning methods are representation-learning methods with multiple levels of representation, obtained by composing simple but non-linear modules that each transform the representation at one level (starting with the raw input) into a representation at a higher, slightly more abstract level. With the composition of enough such transformations, very complex functions can be learned. For classification tasks, higher layers of representation amplify aspects of the input that are important for discrimination and suppress irrelevant variations. An image, for example, comes in the form of an array of pixel values, and the learned features in the first layer of representation typically represent the presence or absence of edges at particular orientations and locations in the image. The second layer typically detects motifs by spotting particular arrangements of edges, regardless of small variations in the edge positions. The third layer may assemble motifs into larger combinations that correspond to parts of familiar objects, and subsequent layers would detect objects as combinations of these parts. The key aspect of deep learning is that these layers of features are not designed by human engineers: they are learned from data using a general-purpose learning procedure. Deep learning is making major advances in solving problems that have resisted the best attempts of the artificial intelligence community for many years. It has turned out to be very good at discovering intricate structures in high-dimensional data and is therefore applicable to many domains of science, business and government. In addition to beating records in image recognition1, 2, 3, 4 and speech recognition5, 6, 7, it has beaten other machine-learning techniques at predicting the activity of potential drug molecules8, analysing particle accelerator data9, 10, reconstructing brain circuits11, and predicting the effects of mutations in non-coding DNA on gene expression and disease12, 13. Perhaps more surprisingly, deep learning has produced extremely promising results for various tasks in natural language understanding14, particularly topic classification, sentiment analysis, question answering15 and language translation16, 17. We think that deep learning will have many more successes in the near future because it requires very little engineering by hand, so it can easily take advantage of increases in the amount of available computation and data. New learning algorithms and architectures that are currently being developed for deep neural networks will only accelerate this progress. The most common form of machine learning, deep or not, is supervised learning. Imagine that we want to build a system that can classify images as containing, say, a house, a car, a person or a pet. We first collect a large data set of images of houses, cars, people and pets, each labelled with its category. During training, the machine is shown an image and produces an output in the form of a vector of scores, one for each category. We want the desired category to have the highest score of all categories, but this is unlikely to happen before training. We compute an objective function that measures the error (or distance) between the output scores and the desired pattern of scores. The machine then modifies its internal adjustable parameters to reduce this error. These adjustable parameters, often called weights, are real numbers that can be seen as 'knobs' that define the input–output function of the machine. In a typical deep-learning system, there may be hundreds of millions of these adjustable weights, and hundreds of millions of labelled examples with which to train the machine. To properly adjust the weight vector, the learning algorithm computes a gradient vector that, for each weight, indicates by what amount the error would increase or decrease if the weight were increased by a tiny amount. The weight vector is then adjusted in the opposite direction to the gradient vector. The objective function, averaged over all the training examples, can be seen as a kind of hilly landscape in the high-dimensional space of weight values. The negative gradient vector indicates the direction of steepest descent in this landscape, taking it closer to a minimum, where the output error is low on average. In practice, most practitioners use a procedure called stochastic gradient descent (SGD). This consists of showing the input vector for a few examples, computing the outputs and the errors, computing the average gradient for those examples, and adjusting the weights accordingly. The process is repeated for many small sets of examples from the training set until the average of the objective function stops decreasing. It is called stochastic because each small set of examples gives a noisy estimate of the average gradient over all examples. This simple procedure usually finds a good set of weights surprisingly quickly when compared with far more elaborate optimization techniques18. After training, the performance of the system is measured on a different set of examples called a test set. This serves to test the generalization ability of the machine — its ability to produce sensible answers on new inputs that it has never seen during training. Many of the current practical applications of machine learning use linear classifiers on top of hand-engineered features. A two-class linear classifier computes a weighted sum of the feature vector components. If the weighted sum is above a threshold, the input is classified as belonging to a particular category. Since the 1960s we have known that linear classifiers can only carve their input space into very simple regions, namely half-spaces separated by a hyperplane19. But problems such as image and speech recognition require the input–output function to be insensitive to irrelevant variations of the input, such as variations in position, orientation or illumination of an object, or variations in the pitch or accent of speech, while being very sensitive to particular minute variations (for example, the difference between a white wolf and a breed of wolf-like white dog called a Samoyed). At the pixel level, images of two Samoyeds in different poses and in different environments may be very different from each other, whereas two images of a Samoyed and a wolf in the same position and on similar backgrounds may be very similar to each other. A linear classifier, or any other 'shallow' classifier operating on raw pixels could not possibly distinguish the latter two, while putting the former two in the same category. This is why shallow classifiers require a good feature extractor that solves the selectivity–invariance dilemma — one that produces representations that are selective to the aspects of the image that are important for discrimination, but that are invariant to irrelevant aspects such as the pose of the animal. To make classifiers more powerful, one can use generic non-linear features, as with kernel methods20, but generic features such as those arising with the Gaussian kernel do not allow the learner to generalize well far from the training examples21. The conventional option is to hand design good feature extractors, which requires a considerable amount of engineering skill and domain expertise. But this can all be avoided if good features can be learned automatically using a general-purpose learning procedure. This is the key advantage of deep learning. A deep-learning architecture is a multilayer stack of simple modules, all (or most) of which are subject to learning, and many of which compute non-linear input–output mappings. Each module in the stack transforms its input to increase both the selectivity and the invariance of the representation. With multiple non-linear layers, say a depth of 5 to 20, a system can implement extremely intricate functions of its inputs that are simultaneously sensitive to minute details — distinguishing Samoyeds from white wolves — and insensitive to large irrelevant variations such as the background, pose, lighting and surrounding objects. From the earliest days of pattern recognition22, 23, the aim of researchers has been to replace hand-engineered features with trainable multilayer networks, but despite its simplicity, the solution was not widely understood until the mid 1980s. As it turns out, multilayer architectures can be trained by simple stochastic gradient descent. As long as the modules are relatively smooth functions of their inputs and of their internal weights, one can compute gradients using the backpropagation procedure. The idea that this could be done, and that it worked, was discovered independently by several different groups during the 1970s and 1980s24, 25, 26, 27. The backpropagation procedure to compute the gradient of an objective function with respect to the weights of a multilayer stack of modules is nothing more than a practical application of the chain rule for derivatives. The key insight is that the derivative (or gradient) of the objective with respect to the input of a module can be computed by working backwards from the gradient with respect to the output of that module (or the input of the subsequent module) (Fig. 1). The backpropagation equation can be applied repeatedly to propagate gradients through all modules, starting from the output at the top (where the network produces its prediction) all the way to the bottom (where the external input is fed). Once these gradients have been computed, it is straightforward to compute the gradients with respect to the weights of each module. Many applications of deep learning use feedforward neural network architectures (Fig. 1), which learn to map a fixed-size input (for example, an image) to a fixed-size output (for example, a probability for each of several categories). To go from one layer to the next, a set of units compute a weighted sum of their inputs from the previous layer and pass the result through a non-linear function. At present, the most popular non-linear function is the rectified linear unit (ReLU), which is simply the half-wave rectifier f(z) = max(z, 0). In past decades, neural nets used smoother non-linearities, such as tanh(z) or 1/(1 + exp(−z)), but the ReLU typically learns much faster in networks with many layers, allowing training of a deep supervised network without unsupervised pre-training28. Units that are not in the input or output layer are conventionally called hidden units. The hidden layers can be seen as distorting the input in a non-linear way so that categories become linearly separable by the last layer (Fig. 1). In the late 1990s, neural nets and backpropagation were largely forsaken by the machine-learning community and ignored by the computer-vision and speech-recognition communities. It was widely thought that learning useful, multistage, feature extractors with little prior knowledge was infeasible. In particular, it was commonly thought that simple gradient descent would get trapped in poor local minima — weight configurations for which no small change would reduce the average error. In practice, poor local minima are rarely a problem with large networks. Regardless of the initial conditions, the system nearly always reaches solutions of very similar quality. Recent theoretical and empirical results strongly suggest that local minima are not a serious issue in general. Instead, the landscape is packed with a combinatorially large number of saddle points where the gradient is zero, and the surface curves up in most dimensions and curves down in the remainder29, 30. The analysis seems to show that saddle points with only a few downward curving directions are present in very large numbers, but almost all of them have very similar values of the objective function. Hence, it does not much matter which of these saddle points the algorithm gets stuck at. Interest in deep feedforward networks was revived around 2006 (refs 31,32,33,34) by a group of researchers brought together by the Canadian Institute for Advanced Research (CIFAR). The researchers introduced unsupervised learning procedures that could create layers of feature detectors without requiring labelled data. The objective in learning each layer of feature detectors was to be able to reconstruct or model the activities of feature detectors (or raw inputs) in the layer below. By 'pre-training' several layers of progressively more complex feature detectors using this reconstruction objective, the weights of a deep network could be initialized to sensible values. A final layer of output units could then be added to the top of the network and the whole deep system could be fine-tuned using standard backpropagation33, 34, 35. This worked remarkably well for recognizing handwritten digits or for detecting pedestrians, especially when the amount of labelled data was very limited36. The first major application of this pre-training approach was in speech recognition, and it was made possible by the advent of fast graphics processing units (GPUs) that were convenient to program37 and allowed researchers to train networks 10 or 20 times faster. In 2009, the approach was used to map short temporal windows of coefficients extracted from a sound wave to a set of probabilities for the various fragments of speech that might be represented by the frame in the centre of the window. It achieved record-breaking results on a standard speech recognition benchmark that used a small vocabulary38 and was quickly developed to give record-breaking results on a large vocabulary task39. By 2012, versions of the deep net from 2009 were being developed by many of the major speech groups6 and were already being deployed in Android phones. For smaller data sets, unsupervised pre-training helps to prevent overfitting40, leading to significantly better generalization when the number of labelled examples is small, or in a transfer setting where we have lots of examples for some 'source' tasks but very few for some 'target' tasks. Once deep learning had been rehabilitated, it turned out that the pre-training stage was only needed for small data sets. There was, however, one particular type of deep, feedforward network that was much easier to train and generalized much better than networks with full connectivity between adjacent layers. This was the convolutional neural network (ConvNet)41, 42. It achieved many practical successes during the period when neural networks were out of favour and it has recently been widely adopted by the computer-vision community. ConvNets are designed to process data that come in the form of multiple arrays, for example a colour image composed of three 2D arrays containing pixel intensities in the three colour channels. Many data modalities are in the form of multiple arrays: 1D for signals and sequences, including language; 2D for images or audio spectrograms; and 3D for video or volumetric images. There are four key ideas behind ConvNets that take advantage of the properties of natural signals: local connections, shared weights, pooling and the use of many layers. The architecture of a typical ConvNet (Fig. 2) is structured as a series of stages. The first few stages are composed of two types of layers: convolutional layers and pooling layers. Units in a convolutional layer are organized in feature maps, within which each unit is connected to local patches in the feature maps of the previous layer through a set of weights called a filter bank. The result of this local weighted sum is then passed through a non-linearity such as a ReLU. All units in a feature map share the same filter bank. Different feature maps in a layer use different filter banks. The reason for this architecture is twofold. First, in array data such as images, local groups of values are often highly correlated, forming distinctive local motifs that are easily detected. Second, the local statistics of images and other signals are invariant to location. In other words, if a motif can appear in one part of the image, it could appear anywhere, hence the idea of units at different locations sharing the same weights and detecting the same pattern in different parts of the array. Mathematically, the filtering operation performed by a feature map is a discrete convolution, hence the name. Although the role of the convolutional layer is to detect local conjunctions of features from the previous layer, the role of the pooling layer is to merge semantically similar features into one. Because the relative positions of the features forming a motif can vary somewhat, reliably detecting the motif can be done by coarse-graining the position of each feature. A typical pooling unit computes the maximum of a local patch of units in one feature map (or in a few feature maps). Neighbouring pooling units take input from patches that are shifted by more than one row or column, thereby reducing the dimension of the representation and creating an invariance to small shifts and distortions. Two or three stages of convolution, non-linearity and pooling are stacked, followed by more convolutional and fully-connected layers. Backpropagating gradients through a ConvNet is as simple as through a regular deep network, allowing all the weights in all the filter banks to be trained. Deep neural networks exploit the property that many natural signals are compositional hierarchies, in which higher-level features are obtained by composing lower-level ones. In images, local combinations of edges form motifs, motifs assemble into parts, and parts form objects. Similar hierarchies exist in speech and text from sounds to phones, phonemes, syllables, words and sentences. The pooling allows representations to vary very little when elements in the previous layer vary in position and appearance. The convolutional and pooling layers in ConvNets are directly inspired by the classic notions of simple cells and complex cells in visual neuroscience43, and the overall architecture is reminiscent of the LGN–V1–V2–V4–IT hierarchy in the visual cortex ventral pathway44. When ConvNet models and monkeys are shown the same picture, the activations of high-level units in the ConvNet explains half of the variance of random sets of 160 neurons in the monkey's inferotemporal cortex45. ConvNets have their roots in the neocognitron46, the architecture of which was somewhat similar, but did not have an end-to-end supervised-learning algorithm such as backpropagation. A primitive 1D ConvNet called a time-delay neural net was used for the recognition of phonemes and simple words47, 48. There have been numerous applications of convolutional networks going back to the early 1990s, starting with time-delay neural networks for speech recognition47 and document reading42. The document reading system used a ConvNet trained jointly with a probabilistic model that implemented language constraints. By the late 1990s this system was reading over 10% of all the cheques in the United States. A number of ConvNet-based optical character recognition and handwriting recognition systems were later deployed by Microsoft49. ConvNets were also experimented with in the early 1990s for object detection in natural images, including faces and hands50, 51, and for face recognition52. Since the early 2000s, ConvNets have been applied with great success to the detection, segmentation and recognition of objects and regions in images. These were all tasks in which labelled data was relatively abundant, such as traffic sign recognition53, the segmentation of biological images54 particularly for connectomics55, and the detection of faces, text, pedestrians and human bodies in natural images36, 50, 51, 56, 57, 58. A major recent practical success of ConvNets is face recognition59. Importantly, images can be labelled at the pixel level, which will have applications in technology, including autonomous mobile robots and self-driving cars60, 61. Companies such as Mobileye and NVIDIA are using such ConvNet-based methods in their upcoming vision systems for cars. Other applications gaining importance involve natural language understanding14 and speech recognition7. Despite these successes, ConvNets were largely forsaken by the mainstream computer-vision and machine-learning communities until the ImageNet competition in 2012. When deep convolutional networks were applied to a data set of about a million images from the web that contained 1,000 different classes, they achieved spectacular results, almost halving the error rates of the best competing approaches1. This success came from the efficient use of GPUs, ReLUs, a new regularization technique called dropout62, and techniques to generate more training examples by deforming the existing ones. This success has brought about a revolution in computer vision; ConvNets are now the dominant approach for almost all recognition and detection tasks4, 58, 59, 63, 64, 65 and approach human performance on some tasks. A recent stunning demonstration combines ConvNets and recurrent net modules for the generation of image captions (Fig. 3). Recent ConvNet architectures have 10 to 20 layers of ReLUs, hundreds of millions of weights, and billions of connections between units. Whereas training such large networks could have taken weeks only two years ago, progress in hardware, software and algorithm parallelization have reduced training times to a few hours. The performance of ConvNet-based vision systems has caused most major technology companies, including Google, Facebook, Microsoft, IBM, Yahoo!, Twitter and Adobe, as well as a quickly growing number of start-ups to initiate research and development projects and to deploy ConvNet-based image understanding products and services. ConvNets are easily amenable to efficient hardware implementations in chips or field-programmable gate arrays66, 67. A number of companies such as NVIDIA, Mobileye, Intel, Qualcomm and Samsung are developing ConvNet chips to enable real-time vision applications in smartphones, cameras, robots and self-driving cars. Deep-learning theory shows that deep nets have two different exponential advantages over classic learning algorithms that do not use distributed representations21. Both of these advantages arise from the power of composition and depend on the underlying data-generating distribution having an appropriate componential structure40. First, learning distributed representations enable generalization to new combinations of the values of learned features beyond those seen during training (for example, 2n combinations are possible with n binary features)68, 69. Second, composing layers of representation in a deep net brings the potential for another exponential advantage70 (exponential in the depth). The hidden layers of a multilayer neural network learn to represent the network's inputs in a way that makes it easy to predict the target outputs. This is nicely demonstrated by training a multilayer neural network to predict the next word in a sequence from a local context of earlier words71. Each word in the context is presented to the network as a one-of-N vector, that is, one component has a value of 1 and the rest are 0. In the first layer, each word creates a different pattern of activations, or word vectors (Fig. 4). In a language model, the other layers of the network learn to convert the input word vectors into an output word vector for the predicted next word, which can be used to predict the probability for any word in the vocabulary to appear as the next word. The network learns word vectors that contain many active components each of which can be interpreted as a separate feature of the word, as was first demonstrated27 in the context of learning distributed representations for symbols. These semantic features were not explicitly present in the input. They were discovered by the learning procedure as a good way of factorizing the structured relationships between the input and output symbols into multiple 'micro-rules'. Learning word vectors turned out to also work very well when the word sequences come from a large corpus of real text and the individual micro-rules are unreliable71. When trained to predict the next word in a news story, for example, the learned word vectors for Tuesday and Wednesday are very similar, as are the word vectors for Sweden and Norway. Such representations are called distributed representations because their elements (the features) are not mutually exclusive and their many configurations correspond to the variations seen in the observed data. These word vectors are composed of learned features that were not determined ahead of time by experts, but automatically discovered by the neural network. Vector representations of words learned from text are now very widely used in natural language applications14, 17, 72, 73, 74, 75, 76. The issue of representation lies at the heart of the debate between the logic-inspired and the neural-network-inspired paradigms for cognition. In the logic-inspired paradigm, an instance of a symbol is something for which the only property is that it is either identical or non-identical to other symbol instances. It has no internal structure that is relevant to its use; and to reason with symbols, they must be bound to the variables in judiciously chosen rules of inference. By contrast, neural networks just use big activity vectors, big weight matrices and scalar non-linearities to perform the type of fast 'intuitive' inference that underpins effortless commonsense reasoning. Before the introduction of neural language models71, the standard approach to statistical modelling of language did not exploit distributed representations: it was based on counting frequencies of occurrences of short symbol sequences of length up to N (called N-grams). The number of possible N-grams is on the order of VN, where V is the vocabulary size, so taking into account a context of more than a handful of words would require very large training corpora. N-grams treat each word as an atomic unit, so they cannot generalize across semantically related sequences of words, whereas neural language models can because they associate each word with a vector of real valued features, and semantically related words end up close to each other in that vector space (Fig. 4). When backpropagation was first introduced, its most exciting use was for training recurrent neural networks (RNNs). For tasks that involve sequential inputs, such as speech and language, it is often better to use RNNs (Fig. 5). RNNs process an input sequence one element at a time, maintaining in their hidden units a 'state vector' that implicitly contains information about the history of all the past elements of the sequence. When we consider the outputs of the hidden units at different discrete time steps as if they were the outputs of different neurons in a deep multilayer network (Fig. 5, right), it becomes clear how we can apply backpropagation to train RNNs. RNNs are very powerful dynamic systems, but training them has proved to be problematic because the backpropagated gradients either grow or shrink at each time step, so over many time steps they typically explode or vanish77, 78. Thanks to advances in their architecture79, 80 and ways of training them81, 82, RNNs have been found to be very good at predicting the next character in the text83 or the next word in a sequence75, but they can also be used for more complex tasks. For example, after reading an English sentence one word at a time, an English 'encoder' network can be trained so that the final state vector of its hidden units is a good representation of the thought expressed by the sentence. This thought vector can then be used as the initial hidden state of (or as extra input to) a jointly trained French 'decoder' network, which outputs a probability distribution for the first word of the French translation. If a particular first word is chosen from this distribution and provided as input to the decoder network it will then output a probability distribution for the second word of the translation and so on until a full stop is chosen17, 72, 76. Overall, this process generates sequences of French words according to a probability distribution that depends on the English sentence. This rather naive way of performing machine translation has quickly become competitive with the state-of-the-art, and this raises serious doubts about whether understanding a sentence requires anything like the internal symbolic expressions that are manipulated by using inference rules. It is more compatible with the view that everyday reasoning involves many simultaneous analogies that each contribute plausibility to a conclusion84, 85. Instead of translating the meaning of a French sentence into an English sentence, one can learn to 'translate' the meaning of an image into an English sentence (Fig. 3). The encoder here is a deep ConvNet that converts the pixels into an activity vector in its last hidden layer. The decoder is an RNN similar to the ones used for machine translation and neural language modelling. There has been a surge of interest in such systems recently (see examples mentioned in ref. 86). RNNs, once unfolded in time (Fig. 5), can be seen as very deep feedforward networks in which all the layers share the same weights. Although their main purpose is to learn long-term dependencies, theoretical and empirical evidence shows that it is difficult to learn to store information for very long78. To correct for that, one idea is to augment the network with an explicit memory. The first proposal of this kind is the long short-term memory (LSTM) networks that use special hidden units, the natural behaviour of which is to remember inputs for a long time79. A special unit called the memory cell acts like an accumulator or a gated leaky neuron: it has a connection to itself at the next time step that has a weight of one, so it copies its own real-valued state and accumulates the external signal, but this self-connection is multiplicatively gated by another unit that learns to decide when to clear the content of the memory. LSTM networks have subsequently proved to be more effective than conventional RNNs, especially when they have several layers for each time step87, enabling an entire speech recognition system that goes all the way from acoustics to the sequence of characters in the transcription. LSTM networks or related forms of gated units are also currently used for the encoder and decoder networks that perform so well at machine translation17, 72, 76. Over the past year, several authors have made different proposals to augment RNNs with a memory module. Proposals include the Neural Turing Machine in which the network is augmented by a 'tape-like' memory that the RNN can choose to read from or write to88, and memory networks, in which a regular network is augmented by a kind of associative memory89. Memory networks have yielded excellent performance on standard question-answering benchmarks. The memory is used to remember the story about which the network is later asked to answer questions. Beyond simple memorization, neural Turing machines and memory networks are being used for tasks that would normally require reasoning and symbol manipulation. Neural Turing machines can be taught 'algorithms'. Among other things, they can learn to output a sorted list of symbols when their input consists of an unsorted sequence in which each symbol is accompanied by a real value that indicates its priority in the list88. Memory networks can be trained to keep track of the state of the world in a setting similar to a text adventure game and after reading a story, they can answer questions that require complex inference90. In one test example, the network is shown a 15-sentence version of the The Lord of the Rings and correctly answers questions such as “where is Frodo now?”89. Unsupervised learning91, 92, 93, 94, 95, 96, 97, 98 had a catalytic effect in reviving interest in deep learning, but has since been overshadowed by the successes of purely supervised learning. Although we have not focused on it in this Review, we expect unsupervised learning to become far more important in the longer term. Human and animal learning is largely unsupervised: we discover the structure of the world by observing it, not by being told the name of every object. Human vision is an active process that sequentially samples the optic array in an intelligent, task-specific way using a small, high-resolution fovea with a large, low-resolution surround. We expect much of the future progress in vision to come from systems that are trained end-to-end and combine ConvNets with RNNs that use reinforcement learning to decide where to look. Systems combining deep learning and reinforcement learning are in their infancy, but they already outperform passive vision systems99 at classification tasks and produce impressive results in learning to play many different video games100. Natural language understanding is another area in which deep learning is poised to make a large impact over the next few years. We expect systems that use RNNs to understand sentences or whole documents will become much better when they learn strategies for selectively attending to one part at a time76, 86. Ultimately, major progress in artificial intelligence will come about through systems that combine representation learning with complex reasoning. Although deep learning and simple reasoning have been used for speech and handwriting recognition for a long time, new paradigms are needed to replace rule-based manipulation of symbolic expressions by operations on large vectors101. Download references The authors would like to thank the Natural Sciences and Engineering Research Council of Canada, the Canadian Institute For Advanced Research (CIFAR), the National Science Foundation and Office of Naval Research for support. Y.L. and Y.B. are CIFAR fellows. Reprints and permissions information is available at www.nature.com/reprints.

Bidirectional recurrent neural networks In the first part of this paper, a regular recurrent neural network (RNN) is extended to a bidirectional recurrent neural network (BRNN). The BRNN can be trained without the limitation of using input information just up to a preset future frame. This is accomplished by training it simultaneously in positive and negative time direction. Structure and training procedure of the proposed network are explained. In regression and classification experiments on artificial data, the proposed structure gives better results than other approaches. For real data, classification experiments for phonemes from the TIMIT database show the same tendency. In the second part of this paper, it is shown how the proposed bidirectional structure can be easily modified to allow efficient estimation of the conditional posterior probability of complete symbol sequences without making any explicit assumption about the shape of the distribution. For this part, experiments on real data are reported

A vector-perturbation technique for near-capacity multiantenna multiuser communication-part I: channel inversion and regularization Recent theoretical results describing the sum capacity when using multiple antennas to communicate with multiple users in a known rich scattering environment have not yet been followed with practical transmission schemes that achieve this capacity. We introduce a simple encoding algorithm that achieves near-capacity at sum rates of tens of bits/channel use. The algorithm is a variation on channel inversion that regularizes the inverse and uses a "sphere encoder" to perturb the data to reduce the power of the transmitted signal. This work is comprised of two parts. In this first part, we show that while the sum capacity grows linearly with the minimum of the number of antennas and users, the sum rate of channel inversion does not. This poor performance is due to the large spread in the singular values of the channel matrix. We introduce regularization to improve the condition of the inverse and maximize the signal-to-interference-plus-noise ratio at the receivers. Regularization enables linear growth and works especially well at low signal-to-noise ratios (SNRs), but as we show in the second part, an additional step is needed to achieve near-capacity performance at all SNRs.

A Nonconservative LMI Condition for Stability of Switched Systems With Guaranteed Dwell Time. Ensuring stability of switched linear systems with a guaranteed dwell time is an important problem in control systems. Several methods have been proposed in the literature to address this problem, but unfortunately they provide sufficient conditions only. This technical note proposes the use of homogeneous polynomial Lyapunov functions in the non-restrictive case where all the subsystems are Hurwitz, showing that a sufficient condition can be provided in terms of an LMI feasibility test by exploiting a key representation of polynomials. Several properties are proved for this condition, in particular that it is also necessary for a sufficiently large degree of these functions. As a result, the proposed condition provides a sequence of upper bounds of the minimum dwell time that approximate it arbitrarily well. Some examples illustrate the proposed approach.

Bessel-Fourier moment-based robust image zero-watermarking iming to resist various signal processing operations and geometric transformations, this paper proposes a robust zero-watermarking algorithm based on a new image moment called Bessel-Fourier moment. First, image normalization is used for the invariance of translation and scaling, then the magnitudes of Bessel-Fourier moments of normalized image are computed, which have rotation invariance and are used to construct the feature image regarded as watermarking. Experimental results and analyses show that the proposed method has strong robustness to various attacks, such as blurring, JPEG compression, Gaussian noise, rotation, scaling, Stirmark and print_photocopy_scan. Compared to the congener zero-watermarking schemes and Zernike moment, the developed method has better performance.

Adaptive Fuzzy Control With Prescribed Performance for Block-Triangular-Structured Nonlinear Systems. In this paper, an adaptive fuzzy control method with prescribed performance is proposed for multi-input and multioutput block-triangular-structured nonlinear systems with immeasurable states. Fuzzy logic systems are adopted to identify the unknown nonlinear system functions. Adaptive fuzzy state observers are designed to solve the problem of unmeasured states, and a new observer-based output-feedb...

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

Efficient and Secure Access Control Scheme in the Standard Model for Vehicular Cloud Computing. Vehicular networking involves the storage, compute, and analysis of massive vehicular data. Vehicular cloud computing, as a special cloud computing platform, seamlessly combines vehicular ad hoc networks and conventional cloud computing. However, in the vehicular cloud computing, there is still the problem of unauthorized users accessing and stealing data. In the traditional ciphertext-policy attribute-based encryption (CP-ABE) scheme, a trusted central authority is employed to manage attributes and distribute keys. Based on multi-authority (MA) CP-ABE, we propose a secure and revocable access control scheme for vehicular cloud computing in this paper, in which the requester can decrypt the ciphertext with only a small amount of computation. We show that our MA-CP-ABE scheme can prevent static corruption of authorities in the standard model under the decisional q-parallel bilinear Diffie-Hellman exponent assumption. Theoretical analysis and experimental simulation results show that our scheme has lower communication cost and lower computational complexity than other schemes.

A Survey of Ant Colony Optimization Based Routing Protocols for Mobile Ad Hoc Networks. Developing highly efficient routing protocols for Mobile Ad hoc NETworks (MANETs) is a challenging task. In order to fulfill multiple routing requirements, such as low packet delay, high packet delivery rate, and effective adaptation to network topology changes with low control overhead, and so on, new ways to approximate solutions to the known NP-hard optimization problem of routing in MANETs have to be investigated. Swarm intelligence (SI)-inspired algorithms have attracted a lot of attention, because they can offer possible optimized solutions ensuring high robustness, flexibility, and low cost. Moreover, they can solve large-scale sophisticated problems without a centralized control entity. A successful example in the SI field is the ant colony optimization (ACO) meta-heuristic. It presents a common framework for approximating solutions to NP-hard optimization problems. ACO has been successfully applied to balance the various routing related requirements in dynamic MANETs. This paper presents a comprehensive survey and comparison of various ACO-based routing protocols in MANETs. The main contributions of this survey include: 1) introducing the ACO principles as applied in routing protocols for MANETs; 2) classifying ACO-based routing approaches reviewed in this paper into five main categories; 3) surveying and comparing the selected routing protocols from the perspective of design and simulation parameters; and 4) discussing open issues and future possible design directions of ACO-based routing protocols.

A Microbial Inspired Routing Protocol for VANETs. We present a bio-inspired unicast routing protocol for vehicular ad hoc networks which uses the cellular attractor selection mechanism to select next hops. The proposed unicast routing protocol based on attractor selecting (URAS) is an opportunistic routing protocol, which is able to change itself adaptively to the complex and dynamic environment by routing feedback packets. We further employ a mu...

Improvement of GPSR Protocol in Vehicular Ad Hoc Network. In a vehicular ad hoc network (VANET), vehicles always move in high-speed which may cause the network topology changes frequently. This is challenging for routing protocols of VANET. Greedy Perimeter Stateless Routing (GPSR) is a representative routing protocol of VANET. However, when constructs routing path, GPSR selects the next hop node which is very easily out of the communication range in greedy forwarding, and builds the path with redundancy in perimeter forwarding. To solve the above-mentioned problems, we proposed Maxduration-Minangle GPSR (MM-GPSR) routing protocol in this paper. In greedy forwarding of MM-GPSR, by defining cumulative communication duration to represent the stability of neighbor nodes, the neighbor node with the maximum cumulative communication duration will be selected as the next hop node. In perimeter forwarding of MM-GPSR when greedy forwarding fails, the concept of minimum angle is introduced as the criterion of the optimal next hop node. By taking the position of neighbor nodes into account and calculating angles formed between neighbors and the destination node, the neighbor node with minimum angle will be selected as the next hop node. By using NS-2 and VanetMobiSim, simulations demonstrate that compared with GPSR, MM-GPSR has obvious improvements in reducing the packet loss rate, decreasing the end-to-end delay and increasing the throughput, and is more suitable for VANET.

Secure Real-Time Traffic Data Aggregation With Batch Verification for Vehicular Cloud in VANETs The vehicular cloud provides many significant advantages to Vehicular ad-hoc Networks (VANETs), such as unlimited storage space, powerful computing capability and timely traffic services. Traffic data aggregation in the vehicular cloud, which can aggregate traffic data from vehicles for further processing and sharing, is very important. Incorrect traffic data feedback may affect traffic safety; therefore, the security of traffic data aggregation should be ensured. In this paper, by using the property of data recovery in the message recovery signature (MRS), we propose a secure real-time traffic data aggregation scheme for vehicular cloud in VANETs. In the proposed scheme, the validity of vehicles’ signatures are verified, and then the original traffic data is recovered from signatures. Moreover, the proposed scheme supports batch verification for multiple vehicles’ signatures. Due to advantages of the MRS, security features such as data confidentiality, privacy preservation and reply attack resistance are preserved. In addition, the comparison and simulation results indicate that the proposed scheme is superior in comparison to previous schemes with respect to the communication and computational cost.

Parking Edge Computing: Parked Vehicle Assisted Task Offloading for Urban VANETs Vehicular edge computing has been a promising paradigm to offer low-latency and high reliability vehicular services for users. Nevertheless, for compute-intensive vehicle applications, most previous researches cannot perform them efficiently due to both the inadequate of infrastructure construction and the computing resource bottleneck of the edge server. Motivated by the fact that there is a larg...

Routing in Flying Ad Hoc Networks: Survey, Constraints, and Future Challenge Perspectives. Owing to the explosive expansion of wireless communication and networking technologies, cost-effective unmanned aerial vehicles (UAVs) have recently emerged and soon they will occupy the major part of our sky. UAVs can be exploited to efficiently accomplish complex missions when cooperatively organized as an ad hoc network, thus creating the well-known flying ad hoc networks (FANETs). The establishment of such networks is not feasible without deploying an efficient networking model allowing a reliable exchange of information between UAVs. FANET inherits common features and characteristics from mobile ad hoc networks (MANETs) and their sub-classes, such as vehicular ad hoc networks (VANETs) and wireless sensor networks (WSNs). Unfortunately, UAVs are often deployed in the sky adopting a mobility model dictated by the nature of missions that they are expected to handle, and therefore, differentiate themselves from any traditional networks. Moreover, several flying constraints and the highly dynamic topology of FANETs make the design of routing protocols a complicated task. In this paper, a comprehensive survey is presented covering the architecture, the constraints, the mobility models, the routing techniques, and the simulation tools dedicated to FANETs. A classification, descriptions, and comparative studies of an important number of existing routing protocols dedicated to FANETs are detailed. Furthermore, the paper depicts future challenge perspectives, helping scientific researchers to discover some themes that have been addressed only ostensibly in the literature and need more investigation. The novelty of this survey is its uniqueness to provide a complete analysis of the major FANET routing protocols and to critically compare them according to different constraints based on crucial parameters, thus better presenting the state of the art of this specific area of research.

Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation Object detection performance, as measured on the canonical PASCAL VOC dataset, has plateaued in the last few years. The best-performing methods are complex ensemble systems that typically combine multiple low-level image features with high-level context. In this paper, we propose a simple and scalable detection algorithm that improves mean average precision (mAP) by more than 30% relative to the previous best result on VOC 2012 -- achieving a mAP of 53.3%. Our approach combines two key insights: (1) one can apply high-capacity convolutional neural networks (CNNs) to bottom-up region proposals in order to localize and segment objects and (2) when labeled training data is scarce, supervised pre-training for an auxiliary task, followed by domain-specific fine-tuning, yields a significant performance boost. Since we combine region proposals with CNNs, we call our method R-CNN: Regions with CNN features. We also present experiments that provide insight into what the network learns, revealing a rich hierarchy of image features. Source code for the complete system is available at http://www.cs.berkeley.edu/~rbg/rcnn.

Adaptive Learning-Based Task Offloading for Vehicular Edge Computing Systems. The vehicular edge computing system integrates the computing resources of vehicles, and provides computing services for other vehicles and pedestrians with task offloading. However, the vehicular task offloading environment is dynamic and uncertain, with fast varying network topologies, wireless channel states, and computing workloads. These uncertainties bring extra challenges to task offloading. In this paper, we consider the task offloading among vehicles, and propose a solution that enables vehicles to learn the offloading delay performance of their neighboring vehicles while offloading computation tasks. We design an adaptive learning based task offloading (ALTO) algorithm based on the multi-armed bandit theory, in order to minimize the average offloading delay. ALTO works in a distributed manner without requiring frequent state exchange, and is augmented with input-awareness and occurrence-awareness to adapt to the dynamic environment. The proposed algorithm is proved to have a sublinear learning regret. Extensive simulations are carried out under both synthetic scenario and realistic highway scenario, and results illustrate that the proposed algorithm achieves low delay performance, and decreases the average delay up to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$30\%$</tex-math></inline-formula> compared with the existing upper confidence bound based learning algorithm.

Visual cryptography for general access structures A visual cryptography scheme for a set P of n participants is a method of encoding a secret image SI into n shadow images called shares, where each participant in P receives one share. Certain qualified subsets of participants can “visually” recover the secret image, but other, forbidden, sets of participants have no information (in an information-theoretic sense) on SI . A “visual” recovery for a set X ⊆ P consists of xeroxing the shares given to the participants in X onto transparencies, and then stacking them. The participants in a qualified set X will be able to see the secret image without any knowledge of cryptography and without performing any cryptographic computation. In this paper we propose two techniques for constructing visual cryptography schemes for general access structures. We analyze the structure of visual cryptography schemes and we prove bounds on the size of the shares distributed to the participants in the scheme. We provide a novel technique for realizing k out of n threshold visual cryptography schemes. Our construction for k out of n visual cryptography schemes is better with respect to pixel expansion than the one proposed by M. Naor and A. Shamir (Visual cryptography, in “Advances in Cryptology—Eurocrypt '94” CA. De Santis, Ed.), Lecture Notes in Computer Science, Vol. 950, pp. 1–12, Springer-Verlag, Berlin, 1995) and for the case of 2 out of n is the best possible. Finally, we consider graph-based access structures, i.e., access structures in which any qualified set of participants contains at least an edge of a given graph whose vertices represent the participants of the scheme.

Secure and privacy preserving keyword searching for cloud storage services Cloud storage services enable users to remotely access data in a cloud anytime and anywhere, using any device, in a pay-as-you-go manner. Moving data into a cloud offers great convenience to users since they do not have to care about the large capital investment in both the deployment and management of the hardware infrastructures. However, allowing a cloud service provider (CSP), whose purpose is mainly for making a profit, to take the custody of sensitive data, raises underlying security and privacy issues. To keep user data confidential against an untrusted CSP, a natural way is to apply cryptographic approaches, by disclosing the data decryption key only to authorized users. However, when a user wants to retrieve files containing certain keywords using a thin client, the adopted encryption system should not only support keyword searching over encrypted data, but also provide high performance. In this paper, we investigate the characteristics of cloud storage services and propose a secure and privacy preserving keyword searching (SPKS) scheme, which allows the CSP to participate in the decipherment, and to return only files containing certain keywords specified by the users, so as to reduce both the computational and communication overhead in decryption for users, on the condition of preserving user data privacy and user querying privacy. Performance analysis shows that the SPKS scheme is applicable to a cloud environment.

Robust and Imperceptible Dual Watermarking for Telemedicine Applications In this paper, the effects of different error correction codes on the robustness and imperceptibility of discrete wavelet transform and singular value decomposition based dual watermarking scheme is investigated. Text and image watermarks are embedded into cover radiological image for their potential application in secure and compact medical data transmission. Four different error correcting codes such as Hamming, the Bose, Ray-Chaudhuri, Hocquenghem (BCH), the Reed---Solomon and hybrid error correcting (BCH and repetition code) codes are considered for encoding of text watermark in order to achieve additional robustness for sensitive text data such as patient identification code. Performance of the proposed algorithm is evaluated against number of signal processing attacks by varying the strength of watermarking and covers image modalities. The experimental results demonstrate that this algorithm provides better robustness without affecting the quality of watermarked image.This algorithm combines the advantages and removes the disadvantages of the two transform techniques. Out of the three error correcting codes tested, it has been found that Reed---Solomon shows the best performance. Further, a hybrid model of two of the error correcting codes (BCH and repetition code) is concatenated and implemented. It is found that the hybrid code achieves better results in terms of robustness. This paper provides a detailed analysis of the obtained experimental results.

Safe mutations for deep and recurrent neural networks through output gradients While neuroevolution (evolving neural networks) has been successful across a variety of domains from reinforcement learning, to artificial life, to evolutionary robotics, it is rarely applied to large, deep neural networks. A central reason is that while random mutation generally works in low dimensions, a random perturbation of thousands or millions of weights will likely break existing functionality. This paper proposes a solution: a family of safe mutation (SM) operators that facilitate exploration without dramatically altering network behavior or requiring additional interaction with the environment. The most effective SM variant scales the degree of mutation of each individual weight according to the sensitivity of the network's outputs to that weight, which requires computing the gradient of outputs with respect to the weights (instead of the gradient of error, as in conventional deep learning). This safe mutation through gradients (SM-G) operator dramatically increases the ability of a simple genetic algorithm-based neuroevolution method to find solutions in high-dimensional domains that require deep and/or recurrent neural networks, including domains that require processing raw pixels. By improving our ability to evolve deep neural networks, this new safer approach to mutation expands the scope of domains amenable to neuroevolution.

Social Robots for (Second) Language Learning in (Migrant) Primary School Children Especially these days, innovation and support from technology to relieve pressure in education is highly urgent. This study tested the potential advantage of a social robot over a tablet in (second) language learning on performance, engagement, and enjoyment. Shortages in primary education call for new technology solutions. Previous studies combined robots with tablets, to compensate for robot’s limitations, however, this study applied direct human–robot interaction. Primary school children (N = 63, aged 4–6) participated in a 3-wave field experiment with story-telling exercises, either with a semi-autonomous robot (without tablet, using WOz) or a tablet. Results showed increased learning gains over time when training with a social robot, compared to the tablet. Children who trained with a robot were more engaged in the story-telling task and enjoyed it more. Robot’s behavioral style (social or neutral) hardly differed overall, however, seems to vary for high versus low educational abilities. While social robots need sophistication before being implemented in schools, our study shows the potential of social robots as tutors in (second) language learning.

Multimodal graph-based reranking for web image search. This paper introduces a web image search reranking approach that explores multiple modalities in a graph-based learning scheme. Different from the conventional methods that usually adopt a single modality or integrate multiple modalities into a long feature vector, our approach can effectively integrate the learning of relevance scores, weights of modalities, and the distance metric and its scaling for each modality into a unified scheme. In this way, the effects of different modalities can be adaptively modulated and better reranking performance can be achieved. We conduct experiments on a large dataset that contains more than 1000 queries and 1 million images to evaluate our approach. Experimental results demonstrate that the proposed reranking approach is more robust than using each individual modality, and it also performs better than many existing methods.

Movie2Comics: Towards a Lively Video Content Presentation a type of artwork, comics is prevalent and popular around the world. However, despite the availability of assistive software and tools, the creation of comics is still a labor-intensive and time-consuming process. This paper proposes a scheme that is able to automatically turn a movie clip to comics. Two principles are followed in the scheme: 1) optimizing the information preservation of the movie; and 2) generating outputs following the rules and the styles of comics. The scheme mainly contains three components: script-face mapping, descriptive picture extraction, and cartoonization. The script-face mapping utilizes face tracking and recognition techniques to accomplish the mapping between characters' faces and their scripts. The descriptive picture extraction then generates a sequence of frames for presentation. Finally, the cartoonization is accomplished via three steps: panel scaling, stylization, and comics layout design. Experiments are conducted on a set of movie clips and the results have demonstrated the usefulness and the effectiveness of the scheme.

View-Based Discriminative Probabilistic Modeling for 3D Object Retrieval and Recognition In view-based 3D object retrieval and recognition, each object is described by multiple views. A central problem is how to estimate the distance between two objects. Most conventional methods integrate the distances of view pairs across two objects as an estimation of their distance. In this paper, we propose a discriminative probabilistic object modeling approach. It builds probabilistic models for each object based on the distribution of its views, and the distance between two objects is defined as the upper bound of the Kullback–Leibler divergence of the corresponding probabilistic models. 3D object retrieval and recognition is accomplished based on the distance measures. We first learn models for each object by the adaptation from a set of global models with a maximum likelihood principle. A further adaption step is then performed to enhance the discriminative ability of the models. We conduct experiments on the ETH 3D object dataset, the National Taiwan University 3D model dataset, and the Princeton Shape Benchmark. We compare our approach with different methods, and experimental results demonstrate the superiority of our approach.

DADNet: Dilated-Attention-Deformable ConvNet for Crowd Counting Most existing CNN-based methods for crowd counting always suffer from large scale variation in objects of interest, leading to density maps of low quality. In this paper, we propose a novel deep model called Dilated-Attention-Deformable ConvNet (DADNet), which consists of two schemes: multi-scale dilated attention and deformable convolutional DME (Density Map Estimation). The proposed model explores a scale-aware attention fusion with various dilation rates to capture different visual granularities of crowd regions of interest, and utilizes deformable convolutions to generate a high-quality density map. There are two merits as follows: (1) varying dilation rates can effectively identify discriminative regions by enlarging the receptive fields of convolutional kernels upon surrounding region cues, and (2) deformable CNN operations promote the accuracy of object localization in the density map by augmenting the spatial object location sampling with adaptive offsets and scalars. DADNet not only excels at capturing rich spatial context of salient and tiny regions of interest simultaneously, but also keeps a robustness to background noises, such as partially occluded objects. Extensive experiments on benchmark datasets verify that DADNet achieves the state-of-the-art performance. Visualization results of the multi-scale attention maps further validate the remarkable interpretability achieved by our solution.

Context-Aware Graph Inference With Knowledge Distillation for Visual Dialog Visual dialog is a challenging task that requires the comprehension of the semantic dependencies among implicit visual and textual contexts. This task can refer to the relational inference in a graphical model with sparse contextual subjects (nodes) and unknown graph structure (relation descriptor); how to model the underlying context-aware relational inference is critical. To this end, we propose a novel context-aware graph (CAG) neural network. We focus on the exploitation of fine-grained relational reasoning with object-level dialog-historical co-reference nodes. The graph structure (relation in dialog) is iteratively updated using an adaptive top- <inline-formula><tex-math notation="LaTeX">$K$</tex-math></inline-formula> message passing mechanism. To eliminate sparse useless relations, each node has dynamic relations in the graph (different related <inline-formula><tex-math notation="LaTeX">$K$</tex-math></inline-formula> neighbor nodes), and only the most relevant nodes are attributive to the context-aware relational graph inference. In addition, to avoid negative performance caused by linguistic bias of history, we propose a pure visual-aware knowledge distillation mechanism named CAG-Distill, in which image-only visual clues are used to regularize the joint dialog-historical contextual awareness at the object-level. Experimental results on VisDial v0.9 and v1.0 datasets show that both CAG and CAG-Distill outperform comparative methods. Visualization results further validate the remarkable interpretability of our graph inference solution.

Mantra-Net: Manipulation Tracing Network For Detection And Localization Of Image Forgeries With Anomalous Features To fight against real-life image forgery, which commonly involves different types and combined manipulations, we propose a unified deep neural architecture called ManTra-Net. Unlike many existing solutions, ManTra-Net is an end-to-end network that performs both detection and localization without extra preprocessing and postprocessing. ManTra-Net is a fully convolutional network and handles images of arbitrary sizes and many known forgery types such splicing, copy-move, removal, enhancement, and even unknown types. This paper has three salient contributions. We design a simple yet effective self-supervised learning task to learn robust image manipulation traces from classifying 385 image manipulation types. Further, we formulate the forgery localization problem as a local anomaly detection problem, design a Z-score feature to capture local anomaly, and propose a novel long short-term memory solution to assess local anomalies. Finally, we carefully conduct ablation experiments to systematically optimize the proposed network design. Our extensive experimental results demonstrate the generalizability, robustness and superiority of ManTra-Net, not only in single types of manipulations/forgeries, but also in their complicated combinations.

A PatchMatch-based Dense-field Algorithm for Video Copy-Move Detection and Localization. We propose a new algorithm for the reliable detection and localization of video copy–move forgeries. Discovering well-crafted video copy–moves may be very difficult, especially when some uniform background is copied to occlude foreground objects. To reliably detect both additive and occlusive copy–moves, we use a dense-field approach, with invariant features that guarantee robustness to several postprocessing operations. To limit complexity, a suitable video-oriented version of PatchMatch is used, with a multiresolution search strategy, and a focus on volumes of interest. Performance assessment relies on a new dataset, designed <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">ad hoc</italic> , with realistic copy–moves and a wide variety of challenging situations. Experimental results show the proposed method to detect and localize video copy–moves with good accuracy even in adverse conditions.

Scalable Person Re-Identification: A Benchmark This paper contributes a new high quality dataset for person re-identification, named \"Market-1501\". Generally, current datasets: 1) are limited in scale, 2) consist of hand-drawn bboxes, which are unavailable under realistic settings, 3) have only one ground truth and one query image for each identity (close environment). To tackle these problems, the proposed Market-1501 dataset is featured in three aspects. First, it contains over 32,000 annotated bboxes, plus a distractor set of over 500K images, making it the largest person re-id dataset to date. Second, images in Market-1501 dataset are produced using the Deformable Part Model (DPM) as pedestrian detector. Third, our dataset is collected in an open system, where each identity has multiple images under each camera. As a minor contribution, inspired by recent advances in large-scale image search, this paper proposes an unsupervised Bag-of-Words descriptor. We view person re-identification as a special task of image search. In experiment, we show that the proposed descriptor yields competitive accuracy on VIPeR, CUHK03, and Market-1501 datasets, and is scalable on the large-scale 500k dataset.

A practical tutorial on the use of nonparametric statistical tests as a methodology for comparing evolutionary and swarm intelligence algorithms The interest in nonparametric statistical analysis has grown recently in the field of computational intelligence. In many experimental studies, the lack of the required properties for a proper application of parametric procedures–independence, normality, and homoscedasticity–yields to nonparametric ones the task of performing a rigorous comparison among algorithms.

A vector-perturbation technique for near-capacity multiantenna multiuser communication-part I: channel inversion and regularization Recent theoretical results describing the sum capacity when using multiple antennas to communicate with multiple users in a known rich scattering environment have not yet been followed with practical transmission schemes that achieve this capacity. We introduce a simple encoding algorithm that achieves near-capacity at sum rates of tens of bits/channel use. The algorithm is a variation on channel inversion that regularizes the inverse and uses a "sphere encoder" to perturb the data to reduce the power of the transmitted signal. This work is comprised of two parts. In this first part, we show that while the sum capacity grows linearly with the minimum of the number of antennas and users, the sum rate of channel inversion does not. This poor performance is due to the large spread in the singular values of the channel matrix. We introduce regularization to improve the condition of the inverse and maximize the signal-to-interference-plus-noise ratio at the receivers. Regularization enables linear growth and works especially well at low signal-to-noise ratios (SNRs), but as we show in the second part, an additional step is needed to achieve near-capacity performance at all SNRs.

A Nonconservative LMI Condition for Stability of Switched Systems With Guaranteed Dwell Time. Ensuring stability of switched linear systems with a guaranteed dwell time is an important problem in control systems. Several methods have been proposed in the literature to address this problem, but unfortunately they provide sufficient conditions only. This technical note proposes the use of homogeneous polynomial Lyapunov functions in the non-restrictive case where all the subsystems are Hurwitz, showing that a sufficient condition can be provided in terms of an LMI feasibility test by exploiting a key representation of polynomials. Several properties are proved for this condition, in particular that it is also necessary for a sufficiently large degree of these functions. As a result, the proposed condition provides a sequence of upper bounds of the minimum dwell time that approximate it arbitrarily well. Some examples illustrate the proposed approach.

Bessel-Fourier moment-based robust image zero-watermarking iming to resist various signal processing operations and geometric transformations, this paper proposes a robust zero-watermarking algorithm based on a new image moment called Bessel-Fourier moment. First, image normalization is used for the invariance of translation and scaling, then the magnitudes of Bessel-Fourier moments of normalized image are computed, which have rotation invariance and are used to construct the feature image regarded as watermarking. Experimental results and analyses show that the proposed method has strong robustness to various attacks, such as blurring, JPEG compression, Gaussian noise, rotation, scaling, Stirmark and print_photocopy_scan. Compared to the congener zero-watermarking schemes and Zernike moment, the developed method has better performance.

Distributed Kalman consensus filter with event-triggered communication: Formulation and stability analysis. •The problem of distributed state estimation in sensor networks with event-triggered communication schedules on both sensor-to-estimator channel and estimator-to-estimator channel is studied.•An event-triggered KCF is designed by deriving the optimal Kalman gain matrix which minimizes the mean squared error.•A computational scalable form of the proposed filter is presented by some approximations.•An appropriate choice of the consensus gain matrix is provided to ensure the stochastic stability of the proposed filter.

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

A robust and flexible digital rights management system for home networks A robust and flexible Digital Rights Management system for home networks is presented. In the proposed system, the central authority delegates its authorization right to the local manager in a home network by issuing a proxy certificate, and the local manager flexibly controls the access rights of home devices on digital contents with its proxy certificate. Furthermore, the proposed system provides a temporary accessing facility for external devices and achieves strong privacy for home devices. For the validation of delegated rights and the revocation of compromised local managers, a hybrid mechanism combining OCSP validation and periodic renewal of proxy certificates is also presented.

An anonymous and secure biometric‐based enterprise digital rights management system for mobile environment Internet-based content distribution facilitates an efficient platform to sell the digital content to the remote users. However, the digital content can be easily copied and redistributed over the network, which causes huge loss to the right holders. On the contrary, the digital rights management (DRM) systems have been introduced in order to regulate authorized content distribution. Enterprise DRM (E-DRM) system is an application of DRM technology, which aims to prevent illegal access of data in an enterprise. Earlier works on E-DRM do not address anonymity, which may lead to identity theft. Recently, Chang et al. proposed an efficient E-DRM mechanism. Their scheme provides greater efficiency and protects anonymity. Unfortunately, we identify that their scheme does not resist the insider attack and password-guessing attack. In addition, Chang et al.'s scheme has some design flaws in the authorization phase. We then point out the requirements of E-DRM system and present the cryptanalysis of Chang et al.'s scheme. In order to remedy the security weaknesses found in Chang et al.'s scheme, we aim to present a secure and efficient E-DRM scheme. The proposed scheme supports the authorized content key distribution and satisfies the desirable security attributes. Additionally, our scheme offers low communication and computation overheads and user's anonymity as well. Through the rigorous formal and informal security analyses, we show that our scheme is secure against possible known attacks. Furthermore, the simulation results for the formal security analysis using the widely accepted Automated Validation of Internet Security Protocols and Applications tool ensure that our scheme is also secure. Copyright (C) 2015 John Wiley & Sons, Ltd.

A Certificateless Authenticated Key Agreement Protocol for Digital Rights Management System.

A more secure digital rights management authentication scheme based on smart card Digital rights management (DRM) system is a technology based mechanism to ensure only authorized access and legal distribution/consumption of the protected digital content. DRM system deals with the whole lifecycle of the digital content including production, management, distribution and consumption. DRM schemes are effective means for the transfer of digital content and safeguard the intellectual property. Recently, Yang et al. proposed a smart-card based DRM authentication scheme providing mutual authentication and session key establishment among all the participants of the DRM environment. We show that their scheme does not resist threats like smart card attack; fails to provide proper password update facility; and does not follow forward secrecy. To overcome these weaknesses, we propose an improvement of Yang et al.’s scheme. The security of our scheme remains intact even if the smart card of the user is lost. In our scheme, user’s smart card is capable of verifying the correctness of the inputted identity and password and hence contributes to achieve an efficient and user- friendly password update phase. In addition, the session keys established between the participating entities are highly secure by virtue of forward secrecy property. We conduct security analysis and comparison with related schemes to evaluate our improved scheme. During comparison, we also highlight the computational cost/time complexity at the user and the server side in terms of the execution time of various operations. The entire analysis shows that the design of the improved scheme is robust enough for the for DRM environment.

Provably Secure Authenticated Content Key Distribution Framework For Iot-Enabled Enterprise Digital Rights Management Systems The internet of things (IoT) is one of the fastest growing technology, which is helping the enterprise industry. However, it brings greater challenges to privacy and security. Moreover, online piracy of content is an emerging issue. Thus, authorised content access is required. Enterprise digital rights management (E-DRM) systems aim to manage the access of electronic records necessitates establishing a standard set of access requirements, binding those access requirements to the electronic records, and computing the permission criteria. In this chain of control, efficient authenticated access to electronic documents is a crucial task. To address the issues, a protocol is designed. Security of the proposed scheme is proved in the random oracle model. Validation of security is done using 'Automated Validation of Internet Security Protocols and Applications (AVISPA)', which indicates that the protocol is safe. Moreover, the comparative study shows that it has a desirable attribute of efficiency.

Enhanced digital rights management authentication scheme based on smart card As a result of the explosive growth in development for computer networks and information technologies in recent years, various activities take place on the Internet, such as the multimedia services. Today, the distribution of large scale digital content (such as audio, video and images) has become easier and more efficient than ever before. However, the intellectual property violation of copyright-protected content has emerged as a major concern. Therefore digital contents are generally encrypted to prevent unauthorised access. A technology of digital rights management (DRM) refers to any of several encryption technologies used to protect digital contents against unauthorised copying, and to control the distribution of the content. Recently, Zhang et al. proposed a DRM authentication scheme based on smart card to realise session key exchange and mutual authentication among all the parties in DRM environment. The proposed scheme is efficient at server side and compact in smart card design. However, the authors will show their proposed scheme cannot resist insider attack and stolen smart card attack, then the authors will propose an improved scheme to preclude above weaknesses.

Distinctive Image Features from Scale-Invariant Keypoints This paper presents a method for extracting distinctive invariant features from images that can be used to perform reliable matching between different views of an object or scene. The features are invariant to image scale and rotation, and are shown to provide robust matching across a substantial range of affine distortion, change in 3D viewpoint, addition of noise, and change in illumination. The features are highly distinctive, in the sense that a single feature can be correctly matched with high probability against a large database of features from many images. This paper also describes an approach to using these features for object recognition. The recognition proceeds by matching individual features to a database of features from known objects using a fast nearest-neighbor algorithm, followed by a Hough transform to identify clusters belonging to a single object, and finally performing verification through least-squares solution for consistent pose parameters. This approach to recognition can robustly identify objects among clutter and occlusion while achieving near real-time performance.

An introduction to ROC analysis Receiver operating characteristics (ROC) graphs are useful for organizing classifiers and visualizing their performance. ROC graphs are commonly used in medical decision making, and in recent years have been used increasingly in machine learning and data mining research. Although ROC graphs are apparently simple, there are some common misconceptions and pitfalls when using them in practice. The purpose of this article is to serve as an introduction to ROC graphs and as a guide for using them in research.

A Comprehensive Survey on Internet of Things (IoT) Toward 5G Wireless Systems Recently, wireless technologies have been growing actively all around the world. In the context of wireless technology, fifth-generation (5G) technology has become a most challenging and interesting topic in wireless research. This article provides an overview of the Internet of Things (IoT) in 5G wireless systems. IoT in the 5G system will be a game changer in the future generation. It will open a door for new wireless architecture and smart services. Recent cellular network LTE (4G) will not be sufficient and efficient to meet the demands of multiple device connectivity and high data rate, more bandwidth, low-latency quality of service (QoS), and low interference. To address these challenges, we consider 5G as the most promising technology. We provide a detailed overview of challenges and vision of various communication industries in 5G IoT systems. The different layers in 5G IoT systems are discussed in detail. This article provides a comprehensive review on emerging and enabling technologies related to the 5G system that enables IoT. We consider the technology drivers for 5G wireless technology, such as 5G new radio (NR), multiple-input–multiple-output antenna with the beamformation technology, mm-wave commutation technology, heterogeneous networks (HetNets), the role of augmented reality (AR) in IoT, which are discussed in detail. We also provide a review on low-power wide-area networks (LPWANs), security challenges, and its control measure in the 5G IoT scenario. This article introduces the role of AR in the 5G IoT scenario. This article also discusses the research gaps and future directions. The focus is also on application areas of IoT in 5G systems. We, therefore, outline some of the important research directions in 5G IoT.

Priced Oblivious Transfer: How to Sell Digital Goods We consider the question of protecting the privacy of customers buying digital goods. More specifically, our goal is to allow a buyer to purchase digital goods from a vendor without letting the vendor learn what, and to the extent possible also when and how much, it is buying. We propose solutions which allow the buyer, after making an initial deposit, to engage in an unlimited number of priced oblivious-transfer protocols, satisfying the following requirements: As long as the buyer's balance contains sufficient funds, it will successfully retrieve the selected item and its balance will be debited by the item's price. However, the buyer should be unable to retrieve an item whose cost exceeds its remaining balance. The vendor should learn nothing except what must inevitably be learned, namely, the amount of interaction and the initial deposit amount (which imply upper bounds on the quantity and total price of all information obtained by the buyer). In particular, the vendor should be unable to learn what the buyer's current balance is or when it actually runs out of its funds. The technical tools we develop, in the process of solving this problem, seem to be of independent interest. In particular, we present the first one-round (two-pass) protocol for oblivious transfer that does not rely on the random oracle model (a very similar protocol was independently proposed by Naor and Pinkas [21]). This protocol is a special case of a more general "conditional disclosure" methodology, which extends a previous approach from [11] and adapts it to the 2-party setting.

Minimum acceleration criterion with constraints implies bang-bang control as an underlying principle for optimal trajectories of arm reaching movements. Rapid arm-reaching movements serve as an excellent test bed for any theory about trajectory formation. How are these movements planned? A minimum acceleration criterion has been examined in the past, and the solution obtained, based on the Euler-Poisson equation, failed to predict that the hand would begin and end the movement at rest (i.e., with zero acceleration). Therefore, this criterion was rejected in favor of the minimum jerk, which was proved to be successful in describing many features of human movements. This letter follows an alternative approach and solves the minimum acceleration problem with constraints using Pontryagin's minimum principle. We use the minimum principle to obtain minimum acceleration trajectories and use the jerk as a control signal. In order to find a solution that does not include nonphysiological impulse functions, constraints on the maximum and minimum jerk values are assumed. The analytical solution provides a three-phase piecewise constant jerk signal (bang-bang control) where the magnitude of the jerk and the two switching times depend on the magnitude of the maximum and minimum available jerk values. This result fits the observed trajectories of reaching movements and takes into account both the extrinsic coordinates and the muscle limitations in a single framework. The minimum acceleration with constraints principle is discussed as a unifying approach for many observations about the neural control of movements.

Wireless Networks with RF Energy Harvesting: A Contemporary Survey Radio frequency (RF) energy transfer and harvesting techniques have recently become alternative methods to power the next generation wireless networks. As this emerging technology enables proactive energy replenishment of wireless devices, it is advantageous in supporting applications with quality of service (QoS) requirements. In this paper, we present a comprehensive literature review on the research progresses in wireless networks with RF energy harvesting capability, referred to as RF energy harvesting networks (RF-EHNs). First, we present an overview of the RF-EHNs including system architecture, RF energy harvesting techniques and existing applications. Then, we present the background in circuit design as well as the state-of-the-art circuitry implementations, and review the communication protocols specially designed for RF-EHNs. We also explore various key design issues in the development of RFEHNs according to the network types, i.e., single-hop networks, multi-antenna networks, relay networks, and cognitive radio networks. Finally, we envision some open research directions.

Automated Detection of Obstructive Sleep Apnea Events from a Single-Lead Electrocardiogram Using a Convolutional Neural Network. In this study, we propose a method for the automated detection of obstructive sleep apnea (OSA) from a single-lead electrocardiogram (ECG) using a convolutional neural network (CNN). A CNN model was designed with six optimized convolution layers including activation, pooling, and dropout layers. One-dimensional (1D) convolution, rectified linear units (ReLU), and max pooling were applied to the convolution, activation, and pooling layers, respectively. For training and evaluation of the CNN model, a single-lead ECG dataset was collected from 82 subjects with OSA and was divided into training (including data from 63 patients with 34,281 events) and testing (including data from 19 patients with 8571 events) datasets. Using this CNN model, a precision of 0.99%, a recall of 0.99%, and an F-score of 0.99% were attained with the training dataset; these values were all 0.96% when the CNN was applied to the testing dataset. These results show that the proposed CNN model can be used to detect OSA accurately on the basis of a single-lead ECG. Ultimately, this CNN model may be used as a screening tool for those suspected to suffer from OSA.

A Muscle Synergy-Driven ANFIS Approach to Predict Continuous Knee Joint Movement Continuous motion prediction plays a significant role in realizing seamless control of robotic exoskeletons and orthoses. Explicitly modeling the relationship between coordinated muscle activations from surface electromyography (sEMG) and human limb movements provides a new path of sEMG-based human–machine interface. Instead of the numeric features from individual channels, we propose a muscle synergy-driven adaptive network-based fuzzy inference system (ANFIS) approach to predict continuous knee joint movements, in which muscle synergy reflects the motor control information to coordinate muscle activations for performing movements. Four human subjects participated in the experiment while walking at five types of speed: 2.0 km/h, 2.5 km/h, 3.0 km/h, 3.5 km/h, and 4.0 km/h. The study finds that the acquired muscle synergies associate the muscle activations with human joint movements in a low-dimensional space and have been further utilized for predicting knee joint angles. The proposed approach outperformed commonly used numeric features from individual sEMG channels with an average correlation coefficient of 0.92 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$ \pm $</tex-math></inline-formula> 0.05. Results suggest that the correlation between muscle activations and knee joint movements is captured by the muscle synergy-driven ANFIS model and can be utilized for the estimation of continuous joint angles.

Robot Gendering: Influences on Trust, Occupational Competency, and Preference of Robot Over Human This paper presents an investigation into the differences found in participants' comfort levels with using new humanoid robots in addition to the tendency to give a gender to a robot designed to be gender neutral. These factors were used to examine participants' perception of occupational competency, trust, and preference for a humanoid robot over a human male or female for various occupations. Our results suggest that comfort level influences these metrics but does not cause a person to ascribe a gender to a gender-neutral robot. These findings suggest that there is no need to perpetuate societal gender norms onto robots. However, even when designing for robot gender neutrality people are still more likely to ascribe a gender to the robot, but this gendering does not significantly impact occupational judgements.

Embodiment and Human-Robot Interaction: A Task-Based Perspective In this work, we further test the hypothesis that physical embodiment has a measurable effect on performance and impression of social interactions. Support for this hypothesis would suggest fundamental differences between virtual agents and robots from a social standpoint and would have significan t implications for human-robot interaction. We have refined our task-based metrics to give a measure- ment, not only of the participant's immediate impressions o f a coach for a task, but also of the participant's performance i n a given task. We measure task performance and participants' impression of a robot's social abilities in a structured tas k based on the Towers of Hanoi puzzle. Our experiment compares aspects of embodiment by evaluating: (1) the difference between a physical robot and a simulated one; and (2) the effect of physical presence through a co-located robot versus a remote, tele-present robot. With a participant pool (n=21) of roboticists and non- roboticists, we were able to show that participants felt that an embodied robot was more appealing and perceptive of the world than non-embodied robots. A larger pool of participants (n=32) also demonstrated that the embodied robot was seen as most helpful, watchful, and enjoyable when compared to a remote tele-present robot and a simulated robot.

Modeling Aspects of Theory of Mind with Markov Random Fields We propose Markov random fields (MRFs) as a probabilistic mathematical model for incorporating the in- ternal states of other agents, both human and robotic, into ro- bot decision making. By using estimates of Theory of Mind (ToM), the mental states of other agents can be incorporated into decision making through statistical inference, allowing robots to balance their own goals and internal objectives with those of other collaborating agents. The MRF model is well-suited to domains in which the joint probability over latent (action) and observed (perceived) variables can be factored into pairwise interactions between these variables. Specifically, these interactions occur through functions that evaluate "local evidence" between an observed and latent variable and "compatibility" between a pair of latent vari- ables. We will describe how experimental findings from the ToM literature can be explained using MRF models, and then show how this framework can be applied to a social robotics task. We will also describe how to use belief prop- agation on a multi-robot MRF as a novel approach to multi- robot coordination, with parallels to human collaboration

An Implemented Theory of Mind to Improve Human-Robot Shared Plans Execution. When a robot has to execute a shared plan with a human, a number of unexpected situations and contingencies can happen due, essentially, to human initiative. For instance, a temporary absence or inattention of the human can entail a partial, and potentially not sufficient, knowledge about the current situation. To ensure a successful and fluent execution of the shared plan the robot might need to detect such situations and be able to provide the information to its human partner about what he missed without being annoying or intrusive. To do so, we have developed a framework which allows a robot to estimate the other agents mental states not only about the environment but also about the state of goals, plans and actions and to take them into account when executing human-robot shared plans.

Deal or No Deal? End-to-End Learning of Negotiation Dialogues. Much of human dialogue occurs in semi-cooperative settings, where agents with different goals attempt to agree on common decisions. Negotiations require complex communication and reasoning skills, but success is easy to measure, making this an interesting task for AI. We gather a large dataset of human-human negotiations on a multi-issue bargaining task, where agents who cannot observe each otheru0027s reward functions must reach an agreement (or a deal) via natural language dialogue. For the first time, we show it is possible to train end-to-end models for negotiation, which must learn both linguistic and reasoning skills with no annotated dialogue states. We also introduce dialogue rollouts, in which the model plans ahead by simulating possible complete continuations of the conversation, and find that this technique dramatically improves performance. Our code and dataset are publicly available (this https URL).

No fair!!: an interaction with a cheating robot Using a humanoid robot and a simple children's game, we examine the degree to which variations in behavior result in attributions of mental state and intentionality. Participants play the well-known children's game ¿rock-paper-scissors¿ against a robot that either plays fairly, or that cheats in one of two ways. In the ¿verbal cheat¿ condition, the robot announces the wrong outcome on several rounds which it loses, declaring itself the winner. In the ¿action cheat¿ condition, the robot changes its gesture after seeing its opponent's play. We find that participants display a greater level of social engagement and make greater attributions of mental state when playing against the robot in the conditions in which it cheats.

Accurate Self-Localization in RFID Tag Information Grids Using FIR Filtering Grid navigation spaces nested with the radio-frequency identification (RFID) tags are promising for industrial and other needs, because each tag can deliver information about a local two-dimensional or three-dimensional surrounding. The approach, however, requires high accuracy in vehicle self-localization. Otherwise, errors may lead to collisions; possibly even fatal. We propose a new extended finite impulse response (EFIR) filtering algorithm and show that it meets this need. The EFIR filter requires an optimal averaging interval, but does not involve the noise statistics which are often not well known to the engineer. It is more accurate than the extended Kalman filter (EKF) under real operation conditions and its iterative algorithm has the Kalman form. Better performance of the proposed EFIR filter is demonstrated based on extensive simulations in a comparison to EKF, which is widely used in RFID tag grids. We also show that errors in noise covariances may provoke divergence in EKF, whereas the EFIR filter remains stable and is thus more robust.

Evolutionary computation: comments on the history and current state Evolutionary computation has started to receive significant attention during the last decade, although the origins can be traced back to the late 1950's. This article surveys the history as well as the current state of this rapidly growing field. We describe the purpose, the general structure, and the working principles of different approaches, including genetic algorithms (GA) (with links to genetic programming (GP) and classifier systems (CS)), evolution strategies (ES), and evolutionary programming (EP) by analysis and comparison of their most important constituents (i.e. representations, variation operators, reproduction, and selection mechanism). Finally, we give a brief overview on the manifold of application domains, although this necessarily must remain incomplete

Supporting social navigation on the World Wide Web This paper discusses a navigation behavior on Internet information services, in particular the World Wide Web, which is characterized by pointing out of information using various communication tools. We call this behavior social navigation as it is based on communication and interaction with other users, be that through email, or any other means of communication. Social navigation phenomena are quite common although most current tools (like Web browsers or email clients) offer very little support for it. We describe why social navigation is useful and how it can be better supported in future systems. We further describe two prototype systems that, although originally not designed explicitly as tools for social navigation, provide features that are typical for social navigation systems. One of these systems, the Juggler system, is a combination of a textual virtual environment and a Web client. The other system is a prototype of a Web- hotlist organizer, called Vortex. We use both systems to describe fundamental principles of social navigation systems.

Proofs of Storage from Homomorphic Identification Protocols Proofs of storage (PoS) are interactive protocols allowing a client to verify that a server faithfully stores a file. Previous work has shown that proofs of storage can be constructed from any homomorphic linear authenticator (HLA). The latter, roughly speaking, are signature/message authentication schemes where `tags' on multiple messages can be homomorphically combined to yield a `tag' on any linear combination of these messages. We provide a framework for building public-key HLAs from any identification protocol satisfying certain homomorphic properties. We then show how to turn any public-key HLA into a publicly-verifiable PoS with communication complexity independent of the file length and supporting an unbounded number of verifications. We illustrate the use of our transformations by applying them to a variant of an identification protocol by Shoup, thus obtaining the first unbounded-use PoS based on factoring (in the random oracle model).

Design, Implementation, and Experimental Results of a Quaternion-Based Kalman Filter for Human Body Motion Tracking Real-time tracking of human body motion is an important technology in synthetic environments, robotics, and other human-computer interaction applications. This paper presents an extended Kalman filter designed for real-time estimation of the orientation of human limb segments. The filter processes data from small inertial/magnetic sensor modules containing triaxial angular rate sensors, accelerometers, and magnetometers. The filter represents rotation using quaternions rather than Euler angles or axis/angle pairs. Preprocessing of the acceleration and magnetometer measurements using the Quest algorithm produces a computed quaternion input for the filter. This preprocessing reduces the dimension of the state vector and makes the measurement equations linear. Real-time implementation and testing results of the quaternion-based Kalman filter are presented. Experimental results validate the filter design, and show the feasibility of using inertial/magnetic sensor modules for real-time human body motion tracking

Switching Stabilization for a Class of Slowly Switched Systems In this technical note, the problem of switching stabilization for slowly switched linear systems is investigated. In particular, the considered systems can be composed of all unstable subsystems. Based on the invariant subspace theory, the switching signal with mode-dependent average dwell time (MDADT) property is designed to exponentially stabilize the underlying system. Furthermore, sufficient condition of stabilization for switched systems with all stable subsystems under MDADT switching is also given. The correctness and effectiveness of the proposed approaches are illustrated by a numerical example.

An efficient scheduling scheme for mobile charger in on-demand wireless rechargeable sensor networks. Existing studies on wireless sensor networks (WSNs) have revealed that the limited battery capacity of sensor nodes (SNs) hinders their perpetual operation. Recent findings in the domain of wireless energy transfer (WET) have attracted a lot of attention of academia and industry to cater the lack of energy in the WSNs. The main idea of WET is to restore the energy of SNs using one or more wireless mobile chargers (MCs), which leads to a new paradigm of wireless rechargeable sensor networks (WRSNs). The determination of an optimal order of charging the SNs (i.e., charging schedule) in an on-demand WRSN is a well-known NP-hard problem. Moreover, care must be taken while designing the charging schedule of an MC as requesting SNs introduce both spatial and temporal constraints. In this paper, we first present a Linear Programming (LP) formulation for the problem of scheduling an MC and then propose an efficient solution based on gravitational search algorithm (GSA). Our method is presented with a novel agent representation scheme and an efficient fitness function. We perform extensive simulations on the proposed scheme to demonstrate its effectiveness over two state-of-the-art algorithms, namely first come first serve (FCFS) and nearest job next with preemption (NJNP). The simulation results reveal that the proposed scheme outperforms both the existing algorithms in terms of charging latency. The virtue of our scheme is also proved by the well-known statistical test, analysis of variance (ANOVA), followed by post hoc analysis.

Hardware Circuits Design and Performance Evaluation of a Soft Lower Limb Exoskeleton Soft lower limb exoskeletons (LLEs) are wearable devices that have good potential in walking rehabilitation and augmentation. While a few studies focused on the structure design and assistance force optimization of the soft LLEs, rarely work has been conducted on the hardware circuits design. The main purpose of this work is to present a new soft LLE for walking efficiency improvement and introduce its hardware circuits design. A soft LLE for hip flexion assistance and a hardware circuits system with scalability were proposed. To assess the efficacy of the soft LLE, the experimental tests that evaluate the sensor data acquisition, force tracking performance, lower limb muscle activity and metabolic cost were conducted. The time error in the peak assistance force was just 1%. The reduction in the normalized root-mean-square EMG of the rectus femoris was 7.1%. The net metabolic cost in exoskeleton on condition was reduced by 7.8% relative to walking with no exoskeleton. The results show that the designed hardware circuits can be applied to the soft LLE and the soft LLE is able to improve walking efficiency of wearers.

Development of a Soft Exosuit for Industrial Applications Wearable robotic devices and exoskeletons, that assist human beings in physically-demanding tasks have the potential to both increase productivity and reduce the risk of musculoskeletal disorders. Soft exoskeletons, known as exosuit, provide improved portability and fit. While many exosuits are populating the market to support light weights, very few are powerful enough to reinforce workers in lifting heavy loads. Adopting the advantages of novel soft-robotic principles, we propose a voice-controlled upper limb exosuit designed to aid its user in lifting up to 10kg per arm. The exosuit uses a wire-driven mechanism to transmit power from a proximally-located actuation stage to the shoulder and elbow. Forces are transmitted to the human body via soft, textile-based components. We evaluate the impact of the device on the muscular effort of a wearer in both a lifting and a holding task. Holding a weight of 14kg with the exosuit results in an average reduction in muscular effort of the biceps brachii and anterior deltoid of 50% and 68%, respectively. Similarly, lifting a weight of 7kg with the exosuit reduces the muscular activity of the same two muscles by 23.4% and 41.2%, respectively.

Exoskeletons for human power augmentation The first load-bearing and energetically autonomous exoskeleton, called the Berkeley Lower Extremity Exoskeleton (BLEEX) walks at the average speed of two miles per hour while carrying 75 pounds of load. The project, funded in 2000 by the Defense Advanced Research Project Agency (DARPA) tackled four fundamental technologies: the exoskeleton architectural design, a control algorithm, a body LAN to host the control algorithm, and an on-board power unit to power the actuators, sensors and the computers. This article gives an overview of the BLEEX project.

Sensing pressure distribution on a lower-limb exoskeleton physical human-machine interface. A sensory apparatus to monitor pressure distribution on the physical human-robot interface of lower-limb exoskeletons is presented. We propose a distributed measure of the interaction pressure over the whole contact area between the user and the machine as an alternative measurement method of human-robot interaction. To obtain this measure, an array of newly-developed soft silicone pressure sensors is inserted between the limb and the mechanical interface that connects the robot to the user, in direct contact with the wearer's skin. Compared to state-of-the-art measures, the advantage of this approach is that it allows for a distributed measure of the interaction pressure, which could be useful for the assessment of safety and comfort of human-robot interaction. This paper presents the new sensor and its characterization, and the development of an interaction measurement apparatus, which is applied to a lower-limb rehabilitation robot. The system is calibrated, and an example its use during a prototypical gait training task is presented.

A soft wearable robotic device for active knee motions using flat pneumatic artificial muscles We present the design of a soft wearable robotic device composed of elastomeric artificial muscle actuators and soft fabric sleeves, for active assistance of knee motions. A key feature of the device is the two-dimensional design of the elastomer muscles that not only allows the compactness of the device, but also significantly simplifies the manufacturing process. In addition, the fabric sleeves make the device lightweight and easily wearable. The elastomer muscles were characterized and demonstrated an initial contraction force of 38N and maximum contraction of 18mm with 104kPa input pressure, approximately. Four elastomer muscles were employed for assisted knee extension and flexion. The robotic device was tested on a 3D printed leg model with an articulated knee joint. Experiments were conducted to examine the relation between systematic change in air pressure and knee extension-flexion. The results showed maximum extension and flexion angles of 95° and 37°, respectively. However, these angles are highly dependent on underlying leg mechanics and positions. The device was also able to generate maximum extension and flexion forces of 3.5N and 7N, respectively.

Robotic Artificial Muscles: Current Progress and Future Perspectives Robotic artificial muscles are a subset of artificial muscles that are capable of producing biologically inspired motions useful for robot systems, i.e., large power-to-weight ratios, inherent compliance, and large range of motions. These actuators, ranging from shape memory alloys to dielectric elastomers, are increasingly popular for biomimetic robots as they may operate without using complex linkage designs or other cumbersome mechanisms. Recent achievements in fabrication, modeling, and control methods have significantly contributed to their potential utilization in a wide range of applications. However, no survey paper has gone into depth regarding considerations pertaining to their selection, design, and usage in generating biomimetic motions. In this paper, we discuss important characteristics and considerations in the selection, design, and implementation of various prominent and unique robotic artificial muscles for biomimetic robots, and provide perspectives on next-generation muscle-powered robots.

Development of muscle suit for upper limb We have been developing a "muscle suit" that provides muscular support to the paralyzed or those otherwise unable to move unaided, as well as to manual workers. The muscle suit is a garment without a metal frame and uses a McKibben actuator driven by compressed air. Because actuators are sewn into the garment, no metal frame is needed, making the muscle suit very light and cheap. With the muscle suit, the patient can willfully control his or her movement. The muscle suit is very helpful for both muscular and emotional support. We propose an armor-type muscle suit in order to overcome issues of a prototype system and then show how abduction motion, which we believe, is the most difficult motion for the upper body, is realized.

Power Assist System HAL-3 for Gait Disorder Person We have developed the power assistive suit, HAL (Hybrid Assistive Leg) which provide the self-walking aid for gait disorder persons or aged persons. In this paper, We introduce HAL-3 system, improving HAL-1,2 systems which had developed previously. EMG signal was used as the input information of power assist controller. We propose a calibration method to identify parameters which relates the EMG to joint torque by using HAL-3. We could obtain suitable torque estimated by EMG and realize an apparatus that enables power to be used for walking and standing up according to the intention of the operator.

Image quality assessment: from error visibility to structural similarity. Objective methods for assessing perceptual image quality traditionally attempted to quantify the visibility of errors (differences) between a distorted image and a reference image using a variety of known properties of the human visual system. Under the assumption that human visual perception is highly adapted for extracting structural information from a scene, we introduce an alternative complementary framework for quality assessment based on the degradation of structural information. As a specific example of this concept, we develop a Structural Similarity Index and demonstrate its promise through a set of intuitive examples, as well as comparison to both subjective ratings and state-of-the-art objective methods on a database of images compressed with JPEG and JPEG2000.

Theory and Experiment on Formation-Containment Control of Multiple Multirotor Unmanned Aerial Vehicle Systems. Formation-containment control problems for multiple multirotor unmanned aerial vehicle (UAV) systems with directed topologies are studied, where the states of leaders form desired formation and the states of followers converge to the convex hull spanned by those of the leaders. First, formation-containment protocols are constructed based on the neighboring information of UAVs. Then, sufficient con...

Response time in man-computer conversational transactions The literature concerning man-computer transactions abounds in controversy about the limits of "system response time" to a user's command or inquiry at a terminal. Two major semantic issues prohibit resolving this controversy. One issue centers around the question of "Response time to what?" The implication is that different human purposes and actions will have different acceptable or useful response times.

Human Shoulder Modeling Including Scapulo-Thoracic Constraint And Joint Sinus Cones In virtual human modeling, the shoulder is usually composed of clavicular, scapular and arm segments related by rotational joints. Although the model is improved, the realistic animation of the shoulder is hardly achieved. This is due to the fact that it is difficult to coordinate the simultaneous motion of the shoulder components in a consistent way. Also, the common use of independent one-degree of freedom (DOF) joint hierarchies does not properly render the 3-D accessibility space of real joints. On the basis of former biomechanical investigations, we propose here an extended shoulder model including scapulo-thoracic constraint and joint sinus cones. As a demonstration, the model is applied, using inverse kinematics, to the animation of a 3-D anatomic muscled skeleton model. (C) 2000 Elsevier Science Ltd. All rights reserved.

Stable fuzzy logic control of a general class of chaotic systems This paper proposes a new approach to the stable design of fuzzy logic control systems that deal with a general class of chaotic processes. The stable design is carried out on the basis of a stability analysis theorem, which employs Lyapunov's direct method and the separate stability analysis of each rule in the fuzzy logic controller (FLC). The stability analysis theorem offers sufficient conditions for the stability of a general class of chaotic processes controlled by Takagi---Sugeno---Kang FLCs. The approach suggested in this paper is advantageous because inserting a new rule requires the fulfillment of only one of the conditions of the stability analysis theorem. Two case studies concerning the fuzzy logic control of representative chaotic systems that belong to the general class of chaotic systems are included in order to illustrate our stable design approach. A set of simulation results is given to validate the theoretical results.

A blind medical image watermarking: DWT-SVD based robust and secure approach for telemedicine applications. In this paper, a blind image watermarking scheme based on discrete wavelet transform (DWT) and singular value decomposition (SVD) is proposed. In this scheme, DWT is applied on ROI (region of interest) of the medical image to get different frequency subbands of its wavelet decomposition. On the low frequency subband LL of the ROI, block-SVD is applied to get different singular matrices. A pair of elements with similar values is identified from the left singular value matrix of these selected blocks. The values of these pairs are modified using certain threshold to embed a bit of watermark content. Appropriate threshold is chosen to achieve the imperceptibility and robustness of medical image and watermark contents respectively. For authentication and identification of original medical image, one watermark image (logo) and other text watermark have been used. The watermark image provides authentication whereas the text data represents electronic patient record (EPR) for identification. At receiving end, blind recovery of both watermark contents is performed by a similar comparison scheme used during the embedding process. The proposed algorithm is applied on various groups of medical images like X-ray, CT scan and mammography. This scheme offers better visibility of watermarked image and recovery of watermark content due to DWT-SVD combination. Moreover, use of Hamming error correcting code (ECC) on EPR text bits reduces the BER and thus provides better recovery of EPR. The performance of proposed algorithm with EPR data coding by Hamming code is compared with the BCH error correcting code and it is found that later one perform better. A result analysis shows that imperceptibility of watermarked image is better as PSNR is above 43 dB and WPSNR is above 52 dB for all set of images. In addition, robustness of the scheme is better than existing scheme for similar set of medical images in terms of normalized correlation coefficient (NCC) and bit-error-rate (BER). An analysis is also carried out to verify the performance of the proposed scheme for different size of watermark contents (image and EPR data). It is observed from analysis that the proposed scheme is also appropriate for watermarking of color image. Using proposed scheme, watermark contents are extracted successfully under various noise attacks like JPEG compression, filtering, Gaussian noise, Salt and pepper noise, cropping, filtering and rotation. Performance comparison of proposed scheme with existing schemes shows proposed scheme has better robustness against different types of attacks. Moreover, the proposed scheme is also robust under set of benchmark attacks known as checkmark attacks.

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

A Dense Phase Descriptor for Human Ear Recognition. Ear print is an imminent biometric modality that has been attracting increasing attention in the biometric community. However, compared with well-established modalities, such as face and fingerprints, a limited number of contributions has been offered on ear imaging. Moreover, only several studies address the aspect of ear characterization (i.e., feature design). In this respect, in this paper, we propose a novel descriptor for ear recognition. The proposed descriptor, namely, dense local phase quantization (DLPQ) is based on the phase responses, which is generated using the well-known LPQ descriptor. Furthermore, local dense histograms are extracted from the horizontal stripes of the phase maps followed by a pooling operation to address viewpoint changes and, finally, concatenated into an ear descriptor. Although the proposed DLPQ descriptor is built on the traditional LPQ, we particularly show that drastic improvements (of over 20%) are attained with respect to this latter descriptor on two benchmark data sets. Furthermore, the proposed descriptor stands out among recent ear descriptors from the literature.

Joint discriminative dimensionality reduction and dictionary learning for face recognition In linear representation based face recognition (FR), it is expected that a discriminative dictionary can be learned from the training samples so that the query sample can be better represented for classification. On the other hand, dimensionality reduction is also an important issue for FR. It cannot only reduce significantly the storage space of face images, but also enhance the discrimination of face feature. Existing methods mostly perform dimensionality reduction and dictionary learning separately, which may not fully exploit the discriminative information in the training samples. In this paper, we propose to learn jointly the projection matrix for dimensionality reduction and the discriminative dictionary for face representation. The joint learning makes the learned projection and dictionary better fit with each other so that a more effective face classification can be obtained. The proposed algorithm is evaluated on benchmark face databases in comparison with existing linear representation based methods, and the results show that the joint learning improves the FR rate, particularly when the number of training samples per class is small.

Sparse Representation Based Fisher Discrimination Dictionary Learning for Image Classification The employed dictionary plays an important role in sparse representation or sparse coding based image reconstruction and classification, while learning dictionaries from the training data has led to state-of-the-art results in image classification tasks. However, many dictionary learning models exploit only the discriminative information in either the representation coefficients or the representation residual, which limits their performance. In this paper we present a novel dictionary learning method based on the Fisher discrimination criterion. A structured dictionary, whose atoms have correspondences to the subject class labels, is learned, with which not only the representation residual can be used to distinguish different classes, but also the representation coefficients have small within-class scatter and big between-class scatter. The classification scheme associated with the proposed Fisher discrimination dictionary learning (FDDL) model is consequently presented by exploiting the discriminative information in both the representation residual and the representation coefficients. The proposed FDDL model is extensively evaluated on various image datasets, and it shows superior performance to many state-of-the-art dictionary learning methods in a variety of classification tasks.

Feature and Rank Level Fusion for Privacy Preserved Multi-Biometric System AbstractPrivacy protection in biometric system is a newly emerging biometric technology that can provide the protection against various attacks by intruders. In this paper, the authors have presented a multi-level of random projection method based on face and ear biometric traits. Privacy preserved templates are used in the proposed system. The main idea behind the privacy preserve computation is the random projection algorithm. Multiple random projection matrixes are used to generate multiple templates for biometric authentication. Newly introduced random fusion method is used in the proposed system; therefore, proposed method can provide better template security, privacy and feature quality. Multiple randomly fused templates are used for recognition purpose and finally decision fusion is applied to generate the final classification result. The proposed method works in a similar way human cognition for face recognition works, furthermore it preserve privacy and multimodality of the system.

Ear recognition using local binary patterns: A comparative experimental study. •A comparative study of ear recognition using local binary patterns variants is done.•A new texture operator is proposed and used as an ear feature descriptor.•Detailed analysis on Identification and verification is conducted separately.•An approximated recognition rate of 99% is achieved by some texture descriptors.•The study has significant insights and can benefit researchers in future works.

Online Palmprint Identification Biometrics-based personal identification is regarded as an effective method for automatically recognizing, with a high confidence, a person's identity. This paper presents a new biometric approach to online personal identification using palmprint technology. In contrast to the existing methods, our online palmprint identification system employs low-resolution palmprint images to achieve effective personal identification. The system consists of two parts: a novel device for online palmprint image acquisition and an efficient algorithm for fast palmprint recognition. A robust image coordinate system is defined to facilitate image alignment for feature extraction. In addition, a 2D Gabor phase encoding scheme is proposed for palmprint feature extraction and representation. The experimental results demonstrate the feasibility of the proposed system.

Improved ear verification after surgery - An approach based on collaborative representation of locally competitive features. •Presents a comprehensive study for biometric verification performance of ears before and after surgery.•Extensive study on different type of ear-surgery is presented along with a new public ear database.•Presents a new feature extraction technique based on Topographic Locally Competitive Algorithm.•Demonstrates superior verification performance on both normal ear database and surgically modified ear database.•Discussion on computational complexity and state-of-art performance.

Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation Object detection performance, as measured on the canonical PASCAL VOC dataset, has plateaued in the last few years. The best-performing methods are complex ensemble systems that typically combine multiple low-level image features with high-level context. In this paper, we propose a simple and scalable detection algorithm that improves mean average precision (mAP) by more than 30% relative to the previous best result on VOC 2012 -- achieving a mAP of 53.3%. Our approach combines two key insights: (1) one can apply high-capacity convolutional neural networks (CNNs) to bottom-up region proposals in order to localize and segment objects and (2) when labeled training data is scarce, supervised pre-training for an auxiliary task, followed by domain-specific fine-tuning, yields a significant performance boost. Since we combine region proposals with CNNs, we call our method R-CNN: Regions with CNN features. We also present experiments that provide insight into what the network learns, revealing a rich hierarchy of image features. Source code for the complete system is available at http://www.cs.berkeley.edu/~rbg/rcnn.

Adaptive Learning-Based Task Offloading for Vehicular Edge Computing Systems. The vehicular edge computing system integrates the computing resources of vehicles, and provides computing services for other vehicles and pedestrians with task offloading. However, the vehicular task offloading environment is dynamic and uncertain, with fast varying network topologies, wireless channel states, and computing workloads. These uncertainties bring extra challenges to task offloading. In this paper, we consider the task offloading among vehicles, and propose a solution that enables vehicles to learn the offloading delay performance of their neighboring vehicles while offloading computation tasks. We design an adaptive learning based task offloading (ALTO) algorithm based on the multi-armed bandit theory, in order to minimize the average offloading delay. ALTO works in a distributed manner without requiring frequent state exchange, and is augmented with input-awareness and occurrence-awareness to adapt to the dynamic environment. The proposed algorithm is proved to have a sublinear learning regret. Extensive simulations are carried out under both synthetic scenario and realistic highway scenario, and results illustrate that the proposed algorithm achieves low delay performance, and decreases the average delay up to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$30\%$</tex-math></inline-formula> compared with the existing upper confidence bound based learning algorithm.

Visual cryptography for general access structures A visual cryptography scheme for a set P of n participants is a method of encoding a secret image SI into n shadow images called shares, where each participant in P receives one share. Certain qualified subsets of participants can “visually” recover the secret image, but other, forbidden, sets of participants have no information (in an information-theoretic sense) on SI . A “visual” recovery for a set X ⊆ P consists of xeroxing the shares given to the participants in X onto transparencies, and then stacking them. The participants in a qualified set X will be able to see the secret image without any knowledge of cryptography and without performing any cryptographic computation. In this paper we propose two techniques for constructing visual cryptography schemes for general access structures. We analyze the structure of visual cryptography schemes and we prove bounds on the size of the shares distributed to the participants in the scheme. We provide a novel technique for realizing k out of n threshold visual cryptography schemes. Our construction for k out of n visual cryptography schemes is better with respect to pixel expansion than the one proposed by M. Naor and A. Shamir (Visual cryptography, in “Advances in Cryptology—Eurocrypt '94” CA. De Santis, Ed.), Lecture Notes in Computer Science, Vol. 950, pp. 1–12, Springer-Verlag, Berlin, 1995) and for the case of 2 out of n is the best possible. Finally, we consider graph-based access structures, i.e., access structures in which any qualified set of participants contains at least an edge of a given graph whose vertices represent the participants of the scheme.

Multi-stage genetic programming: A new strategy to nonlinear system modeling This paper presents a new multi-stage genetic programming (MSGP) strategy for modeling nonlinear systems. The proposed strategy is based on incorporating the individual effect of predictor variables and the interactions among them to provide more accurate simulations. According to the MSGP strategy, an efficient formulation for a problem comprises different terms. In the first stage of the MSGP-based analysis, the output variable is formulated in terms of an influencing variable. Thereafter, the error between the actual and the predicted value is formulated in terms of a new variable. Finally, the interaction term is derived by formulating the difference between the actual values and the values predicted by the individually developed terms. The capabilities of MSGP are illustrated by applying it to the formulation of different complex engineering problems. The problems analyzed herein include the following: (i) simulation of pH neutralization process, (ii) prediction of surface roughness in end milling, and (iii) classification of soil liquefaction conditions. The validity of the proposed strategy is confirmed by applying the derived models to the parts of the experimental results that were not included in the analyses. Further, the external validation of the models is verified using several statistical criteria recommended by other researchers. The MSGP-based solutions are capable of effectively simulating the nonlinear behavior of the investigated systems. The results of MSGP are found to be more accurate than those of standard GP and artificial neural network-based models.

Collaborative Mobile Charging The limited battery capacity of sensor nodes has become one of the most critical impediments that stunt the deployment of wireless sensor networks (WSNs). Recent breakthroughs in wireless energy transfer and rechargeable lithium batteries provide a promising alternative to power WSNs: mobile vehicles/robots carrying high volume batteries serve as mobile chargers to periodically deliver energy to sensor nodes. In this paper, we consider how to schedule multiple mobile chargers to optimize energy usage effectiveness, such that every sensor will not run out of energy. We introduce a novel charging paradigm, collaborative mobile charging, where mobile chargers are allowed to intentionally transfer energy between themselves. To provide some intuitive insights into the problem structure, we first consider a scenario that satisfies three conditions, and propose a scheduling algorithm, PushWait, which is proven to be optimal and can cover a one-dimensional WSN of infinite length. Then, we remove the conditions one by one, investigating chargers' scheduling in a series of scenarios ranging from the most restricted one to a general 2D WSN. Through theoretical analysis and simulations, we demonstrate the advantages of the proposed algorithms in energy usage effectiveness and charging coverage.

A blind medical image watermarking: DWT-SVD based robust and secure approach for telemedicine applications. In this paper, a blind image watermarking scheme based on discrete wavelet transform (DWT) and singular value decomposition (SVD) is proposed. In this scheme, DWT is applied on ROI (region of interest) of the medical image to get different frequency subbands of its wavelet decomposition. On the low frequency subband LL of the ROI, block-SVD is applied to get different singular matrices. A pair of elements with similar values is identified from the left singular value matrix of these selected blocks. The values of these pairs are modified using certain threshold to embed a bit of watermark content. Appropriate threshold is chosen to achieve the imperceptibility and robustness of medical image and watermark contents respectively. For authentication and identification of original medical image, one watermark image (logo) and other text watermark have been used. The watermark image provides authentication whereas the text data represents electronic patient record (EPR) for identification. At receiving end, blind recovery of both watermark contents is performed by a similar comparison scheme used during the embedding process. The proposed algorithm is applied on various groups of medical images like X-ray, CT scan and mammography. This scheme offers better visibility of watermarked image and recovery of watermark content due to DWT-SVD combination. Moreover, use of Hamming error correcting code (ECC) on EPR text bits reduces the BER and thus provides better recovery of EPR. The performance of proposed algorithm with EPR data coding by Hamming code is compared with the BCH error correcting code and it is found that later one perform better. A result analysis shows that imperceptibility of watermarked image is better as PSNR is above 43 dB and WPSNR is above 52 dB for all set of images. In addition, robustness of the scheme is better than existing scheme for similar set of medical images in terms of normalized correlation coefficient (NCC) and bit-error-rate (BER). An analysis is also carried out to verify the performance of the proposed scheme for different size of watermark contents (image and EPR data). It is observed from analysis that the proposed scheme is also appropriate for watermarking of color image. Using proposed scheme, watermark contents are extracted successfully under various noise attacks like JPEG compression, filtering, Gaussian noise, Salt and pepper noise, cropping, filtering and rotation. Performance comparison of proposed scheme with existing schemes shows proposed scheme has better robustness against different types of attacks. Moreover, the proposed scheme is also robust under set of benchmark attacks known as checkmark attacks.

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

A Tutorial on Ultrareliable and Low-Latency Communications in 6G: Integrating Domain Knowledge Into Deep Learning As one of the key communication scenarios in the fifth-generation and also the sixth-generation (6G) mobile communication networks, ultrareliable and low-latency communications (URLLCs) will be central for the development of various emerging mission-critical applications. State-of-the-art mobile communication systems do not fulfill the end-to-end delay and overall reliability requirements of URLLCs. In particular, a holistic framework that takes into account latency, reliability, availability, scalability, and decision-making under uncertainty is lacking. Driven by recent breakthroughs in deep neural networks, deep learning algorithms have been considered as promising ways of developing enabling technologies for URLLCs in future 6G networks. This tutorial illustrates how domain knowledge (models, analytical tools, and optimization frameworks) of communications and networking can be integrated into different kinds of deep learning algorithms for URLLCs. We first provide some background of URLLCs and review promising network architectures and deep learning frameworks for 6G. To better illustrate how to improve learning algorithms with domain knowledge, we revisit model-based analytical tools and cross-layer optimization frameworks for URLLCs. Following this, we examine the potential of applying supervised/unsupervised deep learning and deep reinforcement learning in URLLCs and summarize related open problems. Finally, we provide simulation and experimental results to validate the effectiveness of different learning algorithms and discuss future directions.

A MTC traffic generation and QCI priority-first scheduling algorithm over LTE As (M2M) Machine-To-Machine, communication continues to grow rapidly, a full study on overload control approach to manage the data and signaling of H2H traffic from massive MTC devices is required. In this paper, a new M2M resource-scheduling algorithm for Long Term Evolution (LTE) is proposed. It provides Quality of Service (QoS) guarantee to Guaranteed Bit Rate (GBR) services, we set priorities for the critical M2M services to guarantee the transportation of GBR services, which have high QoS needs. Additionally, we simulate and compare different methods and offer further observations on the solution design.

On Service Resilience in Cloud-Native 5G Mobile Systems. To cope with the tremendous growth in mobile data traffic on one hand, and the modest average revenue per user on the other hand, mobile operators have been exploring network virtualization and cloud computing technologies to build cost-efficient and elastic mobile networks and to have them offered as a cloud service. In such cloud-based mobile networks, ensuring service resilience is an important challenge to tackle. Indeed, high availability and service reliability are important requirements of carrier grade, but not necessarily intrinsic features of cloud computing. Building a system that requires the five nines reliability on a platform that may not always grant it is, therefore, a hurdle. Effectively, in carrier cloud, service resilience can be heavily impacted by a failure of any network function (NF) running on a virtual machine (VM). In this paper, we introduce a framework, along with efficient and proactive restoration mechanisms, to ensure service resilience in carrier cloud. As restoration of a NF failure impacts a potential number of users, adequate network overload control mechanisms are also proposed. A mathematical model is developed to evaluate the performance of the proposed mechanisms. The obtained results are encouraging and demonstrate that the proposed mechanisms efficiently achieve their design goals.

Methodology for the Design and Evaluation of Self-Healing LTE Networks. Self-healing networks aim to detect cells with service degradation, identify the fault cause of their problem, and execute compensation and repair actions. The development of this type of automatic system presents several challenges to be confronted. The first challenge is the scarce number of historically reported faults, which greatly complicates the evaluation of novel self-healing techniques. ...

Multi-agent deep reinforcement learning: a survey The advances in reinforcement learning have recorded sublime success in various domains. Although the multi-agent domain has been overshadowed by its single-agent counterpart during this progress, multi-agent reinforcement learning gains rapid traction, and the latest accomplishments address problems with real-world complexity. This article provides an overview of the current developments in the field of multi-agent deep reinforcement learning. We focus primarily on literature from recent years that combines deep reinforcement learning methods with a multi-agent scenario. To survey the works that constitute the contemporary landscape, the main contents are divided into three parts. First, we analyze the structure of training schemes that are applied to train multiple agents. Second, we consider the emergent patterns of agent behavior in cooperative, competitive and mixed scenarios. Third, we systematically enumerate challenges that exclusively arise in the multi-agent domain and review methods that are leveraged to cope with these challenges. To conclude this survey, we discuss advances, identify trends, and outline possible directions for future work in this research area.

Network Slicing and Softwarization: A Survey on Principles, Enabling Technologies, and Solutions. Network slicing has been identified as the backbone of the rapidly evolving 5G technology. However, as its consolidation and standardization progress, there are no literatures that comprehensively discuss its key principles, enablers, and research challenges. This paper elaborates network slicing from an end-to-end perspective detailing its historical heritage, principal concepts, enabling technol...

Computational thinking Summary form only given. My vision for the 21st century, Computational Thinking, will be a fundamental skill used by everyone in the world. To reading, writing, and arithmetic, we should add computational thinking to every child's analytical ability. Computational thinking involves solving problems, designing systems, and understanding human behavior by drawing on the concepts fundamental to computer science. Thinking like a computer scientist means more than being able to program a computer. It requires the ability to abstract and thus to think at multiple levels of abstraction. In this talk I will give many examples of computational thinking, argue that it has already influenced other disciplines, and promote the idea that teaching computational thinking can not only inspire future generations to enter the field of computer science but benefit people in all fields.

Adam: A Method for Stochastic Optimization. We introduce Adam, an algorithm for first-order gradient-based optimization of stochastic objective functions, based on adaptive estimates of lower-order moments. The method is straightforward to implement, is computationally efficient, has little memory requirements, is invariant to diagonal rescaling of the gradients, and is well suited for problems that are large in terms of data and/or parameters. The method is also appropriate for non-stationary objectives and problems with very noisy and/or sparse gradients. The hyper-parameters have intuitive interpretations and typically require little tuning. Some connections to related algorithms, on which Adam was inspired, are discussed. We also analyze the theoretical convergence properties of the algorithm and provide a regret bound on the convergence rate that is comparable to the best known results under the online convex optimization framework. Empirical results demonstrate that Adam works well in practice and compares favorably to other stochastic optimization methods. Finally, we discuss AdaMax, a variant of Adam based on the infinity norm.

Untangling Blockchain: A Data Processing View of Blockchain Systems. Blockchain technologies are gaining massive momentum in the last few years. Blockchains are distributed ledgers that enable parties who do not fully trust each other to maintain a set of global states. The parties agree on the existence, values, and histories of the states. As the technology landscape is expanding rapidly, it is both important and challenging to have a firm grasp of what the core ...

Multivariate Short-Term Traffic Flow Forecasting Using Time-Series Analysis Existing time-series models that are used for short-term traffic condition forecasting are mostly univariate in nature. Generally, the extension of existing univariate time-series models to a multivariate regime involves huge computational complexities. A different class of time-series models called structural time-series model (STM) (in its multivariate form) has been introduced in this paper to develop a parsimonious and computationally simple multivariate short-term traffic condition forecasting algorithm. The different components of a time-series data set such as trend, seasonal, cyclical, and calendar variations can separately be modeled in STM methodology. A case study at the Dublin, Ireland, city center with serious traffic congestion is performed to illustrate the forecasting strategy. The results indicate that the proposed forecasting algorithm is an effective approach in predicting real-time traffic flow at multiple junctions within an urban transport network.

Dynamic transfer among alternative controllers and its relation to antiwindup controller design Advanced control strategies and modern consulting provide new challenges for the classical problem of bumpless transfer. It can, for example, be necessary to transfer between an only approximately known existing analog controller and a new digital or adaptive controller without accessing any states. Transfer ought to be bidirectional and not presuppose steady state, so that an immediate back-transfer is possible if the new controller should drive the plant unstable. We present a scheme that meets these requirements. By casting the problem of bidirectional transfer into an associated tracking control problem, systematic analysis and design procedures from control theory can be applied. The associated control problem also has a correspondence to the design of antiwindup controllers. The paper includes laboratory and industrial applications.

Adaptive dynamic programming and optimal control of nonlinear nonaffine systems. In this paper, a novel optimal control design scheme is proposed for continuous-time nonaffine nonlinear dynamic systems with unknown dynamics by adaptive dynamic programming (ADP). The proposed methodology iteratively updates the control policy online by using the state and input information without identifying the system dynamics. An ADP algorithm is developed, and can be applied to a general class of nonlinear control design problems. The convergence analysis for the designed control scheme is presented, along with rigorous stability analysis for the closed-loop system. The effectiveness of this new algorithm is illustrated by two simulation examples.

Adaptive fuzzy tracking control for switched uncertain strict-feedback nonlinear systems. •Adaptive tracking control for switched strict-feedback nonlinear systems is proposed.•The generalized fuzzy hyperbolic model is used to approximate nonlinear functions.•The designed controller has fewer design parameters comparing with existing methods.

Learning Feature Recovery Transformer for Occluded Person Re-Identification One major issue that challenges person re-identification (Re-ID) is the ubiquitous occlusion over the captured persons. There are two main challenges for the occluded person Re-ID problem, i.e., the interference of noise during feature matching and the loss of pedestrian information brought by the occlusions. In this paper, we propose a new approach called Feature Recovery Transformer (FRT) to address the two challenges simultaneously, which mainly consists of visibility graph matching and feature recovery transformer. To reduce the interference of the noise during feature matching, we mainly focus on visible regions that appear in both images and develop a visibility graph to calculate the similarity. In terms of the second challenge, based on the developed graph similarity, for each query image, we propose a recovery transformer that exploits the feature sets of its k-nearest neighbors in the gallery to recover the complete features. Extensive experiments across different person Re-ID datasets, including occluded, partial and holistic datasets, demonstrate the effectiveness of FRT. Specifically, FRT significantly outperforms state-of-the-art results by at least 6.2% Rank- 1 accuracy and 7.2% mAP scores on the challenging Occluded-Duke dataset.

Robust circularly orthogonal moment based on Chebyshev rational function. The circularly orthogonal moments have been widely used in many computer vision applications. Unfortunately, they suffer from two errors namely numerical integration error and geometric error, which heavily degrade their reconstruction accuracy and pattern recognition performance. This paper describes a new kind of circularly orthogonal moments based on Chebyshev rational function. Unlike the conventional circularly orthogonal moments which have been defined in a unit disk, the proposed moment is defined in whole polar coordinates domain. In addition, given an order n, its radial projection function is smoother and oscillates at lower frequency compared with the existing circularly orthogonal moments, and so it is free of the geometric error and highly robust to the numerical integration error. Experimental results indicate that the proposed moments perform better in image reconstruction and pattern classification, and yield higher tolerance to image noise and smooth distortion in comparison with the existing circularly orthogonal moments.

Combined invariants to similarity transformation and to blur using orthogonal Zernike moments. The derivation of moment invariants has been extensively investigated in the past decades. In this paper, we construct a set of invariants derived from Zernike moments which is simultaneously invariant to similarity transformation and to convolution with circularly symmetric point spread function (PSF). Two main contributions are provided: the theoretical framework for deriving the Zernike moments of a blurred image and the way to construct the combined geometric-blur invariants. The performance of the proposed descriptors is evaluated with various PSFs and similarity transformations. The comparison of the proposed method with the existing ones is also provided in terms of pattern recognition accuracy, template matching and robustness to noise. Experimental results show that the proposed descriptors perform on the overall better.

Fast computation of Jacobi-Fourier moments for invariant image recognition The Jacobi-Fourier moments (JFMs) provide a wide class of orthogonal rotation invariant moments (ORIMs) which are useful for many image processing, pattern recognition and computer vision applications. They, however, suffer from high time complexity and numerical instability at high orders of moment. In this paper, a fast method based on the recursive computation of radial kernel function of JFMs is proposed which not only reduces time complexity but also improves their numerical stability. Fast recursive method for the computation of Jacobi-Fourier moments is proposed.The proposed method not only reduces time complexity but also improves numerical stability of moments.Better image reconstruction is achieved with lower reconstruction error.Proposed method is useful for many image processing, pattern recognition and computer vision applications.

Radial shifted Legendre moments for image analysis and invariant image recognition. The rotation, scaling and translation invariant property of image moments has a high significance in image recognition. Legendre moments as a classical orthogonal moment have been widely used in image analysis and recognition. Since Legendre moments are defined in Cartesian coordinate, the rotation invariance is difficult to achieve. In this paper, we first derive two types of transformed Legendre polynomial: substituted and weighted radial shifted Legendre polynomials. Based on these two types of polynomials, two radial orthogonal moments, named substituted radial shifted Legendre moments and weighted radial shifted Legendre moments (SRSLMs and WRSLMs) are proposed. The proposed moments are orthogonal in polar coordinate domain and can be thought as generalized and orthogonalized complex moments. They have better image reconstruction performance, lower information redundancy and higher noise robustness than the existing radial orthogonal moments. At last, a mathematical framework for obtaining the rotation, scaling and translation invariants of these two types of radial shifted Legendre moments is provided. Theoretical and experimental results show the superiority of the proposed methods in terms of image reconstruction capability and invariant recognition accuracy under both noisy and noise-free conditions.

Lossless medical image watermarking method based on significant difference of cellular automata transform coefficient. Conventional medical image watermarking techniques focus on improving invisibility and robustness of the watermarking mechanism to prevent medical disputes. This paper proposes a medical image watermarking algorithm based on the significant difference of cellular automata transform (CAT) for copyright protection. The medical image is firstly subsampled into four subimages, and two images are randomly chosen to obtain two low-frequency bandwidths using CAT. Coefficients within a low-frequency bandwidth are important information in an image. Hence, the difference between two low-frequency bandwidths is used as an important feature in the medical image. From these important features, watermarks and cover images can be used to generate an ownership share image (OSI) used for verifying the medical image. Besides appearing like cover images, the OSI will also be able to register with a third party. When a suspected medical image requires verification, the important features from the suspected medical image are first extracted. The master share image (MSI) can be generated from the important features from the suspected medical image. Lastly, the OSI and MSI can be combined to extract the watermark to verify the suspected medical image. The advantage of our proposed method is that the medical image does not require alteration to protect the copyright of the medical image. This means that while the image is protected, medical disputes will be unlikely and the appearance of the registered OSI will carry significant data to make management more convenient. Lastly, the proposed method has the features of having better security, invisibility, and robustness. Moreover, experimental results have demonstrated that our method results in good performance.

Robust watermarking scheme for color image based on quaternion-type moment invariants and visual cryptography. This paper introduces a novel robust watermarking scheme for copyright protection of color image based on quaternion-type moment invariants and visual cryptography. As a secure way to allow secret sharing of images, visual cryptography realizes encryption of classified information and the decryption is performed through human visual system. The proposed scheme represents the color image into a quaternion matrix, so that it can deal with the multichannel information in a holistic way. Then the quaternion moments are applied to extract the invariant features, which are crucial to generate the master share. Together with the scrambled watermark, they are used for constructing the ownership share based on visual cryptography. Afterwards, the ownership share is registered and responsible for authentication. A set of experiments has been conducted to illustrate the validity and feasibility of the proposed scheme as well as better robustness against different attacks.

Two-Dimensional Polar Harmonic Transforms for Invariant Image Representation This paper introduces a set of 2D transforms, based on a set of orthogonal projection bases, to generate a set of features which are invariant to rotation. We call these transforms Polar Harmonic Transforms (PHTs). Unlike the well-known Zernike and pseudo-Zernike moments, the kernel computation of PHTs is extremely simple and has no numerical stability issue whatsoever. This implies that PHTs encompass the orthogonality and invariance advantages of Zernike and pseudo-Zernike moments, but are free from their inherent limitations. This also means that PHTs are well suited for application where maximal discriminant information is needed. Furthermore, PHTs make available a large set of features for further feature selection in the process of seeking for the best discriminative or representative features for a particular application.

Tabu Search - Part I

Joint Optimization of Radio and Computational Resources for Multicell Mobile-Edge Computing Migrating computational intensive tasks from mobile devices to more resourceful cloud servers is a promising technique to increase the computational capacity of mobile devices while saving their battery energy. In this paper, we consider a MIMO multicell system where multiple mobile users (MUs) ask for computation offloading to a common cloud server. We formulate the offloading problem as the joint optimization of the radio resources􀀀the transmit precoding matrices of the MUs􀀀and the computational resources􀀀the CPU cycles/second assigned by the cloud to each MU􀀀in order to minimize the overall users’ energy consumption, while meeting latency constraints. The resulting optimization problem is nonconvex (in the objective function and constraints). Nevertheless, in the single-user case, we are able to compute the global optimal solution in closed form. In the more challenging multiuser scenario, we propose an iterative algorithm, based on a novel successive convex approximation technique, converging to a local optimal solution of the original nonconvex problem. We then show that the proposed algorithmic framework naturally leads to a distributed and parallel implementation across the radio access points, requiring only a limited coordination/signaling with the cloud. Numerical results show that the proposed schemes outperform disjoint optimization algorithms.

Symbolic model checking for real-time systems We describe finite-state programs over real-numbered time in a guarded-command language with real-valued clocks or, equivalently, as finite automata with real-valued clocks. Model checking answers the question which states of a real-time program satisfy a branching-time specification (given in an extension of CTL with clock variables). We develop an algorithm that computes this set of states symbolically as a fixpoint of a functional on state predicates, without constructing the state space. For this purpose, we introduce a μ-calculus on computation trees over real-numbered time. Unfortunately, many standard program properties, such as response for all nonzeno execution sequences (during which time diverges), cannot be characterized by fixpoints: we show that the expressiveness of the timed μ-calculus is incomparable to the expressiveness of timed CTL. Fortunately, this result does not impair the symbolic verification of "implementable" real-time programs-those whose safety constraints are machine-closed with respect to diverging time and whose fairness constraints are restricted to finite upper bounds on clock values. All timed CTL properties of such programs are shown to be computable as finitely approximable fixpoints in a simple decidable theory.

Teaching-Learning-Based Optimization: An optimization method for continuous non-linear large scale problems An efficient optimization method called 'Teaching-Learning-Based Optimization (TLBO)' is proposed in this paper for large scale non-linear optimization problems for finding the global solutions. The proposed method is based on the effect of the influence of a teacher on the output of learners in a class. The basic philosophy of the method is explained in detail. The effectiveness of the method is tested on many benchmark problems with different characteristics and the results are compared with other population based methods.

An Automatic Screening Approach for Obstructive Sleep Apnea Diagnosis Based on Single-Lead Electrocardiogram Traditional approaches for obstructive sleep apnea (OSA) diagnosis are apt to using multiple channels of physiological signals to detect apnea events by dividing the signals into equal-length segments, which may lead to incorrect apnea event detection and weaken the performance of OSA diagnosis. This paper proposes an automatic-segmentation-based screening approach with the single channel of Electrocardiogram (ECG) signal for OSA subject diagnosis, and the main work of the proposed approach lies in three aspects: (i) an automatic signal segmentation algorithm is adopted for signal segmentation instead of the equal-length segmentation rule; (ii) a local median filter is improved for reduction of the unexpected RR intervals before signal segmentation; (iii) the designed OSA severity index and additional admission information of OSA suspects are plugged into support vector machine (SVM) for OSA subject diagnosis. A real clinical example from PhysioNet database is provided to validate the proposed approach and an average accuracy of 97.41% for subject diagnosis is obtained which demonstrates the effectiveness for OSA diagnosis.

Automated Detection of Obstructive Sleep Apnea Events from a Single-Lead Electrocardiogram Using a Convolutional Neural Network. In this study, we propose a method for the automated detection of obstructive sleep apnea (OSA) from a single-lead electrocardiogram (ECG) using a convolutional neural network (CNN). A CNN model was designed with six optimized convolution layers including activation, pooling, and dropout layers. One-dimensional (1D) convolution, rectified linear units (ReLU), and max pooling were applied to the convolution, activation, and pooling layers, respectively. For training and evaluation of the CNN model, a single-lead ECG dataset was collected from 82 subjects with OSA and was divided into training (including data from 63 patients with 34,281 events) and testing (including data from 19 patients with 8571 events) datasets. Using this CNN model, a precision of 0.99%, a recall of 0.99%, and an F-score of 0.99% were attained with the training dataset; these values were all 0.96% when the CNN was applied to the testing dataset. These results show that the proposed CNN model can be used to detect OSA accurately on the basis of a single-lead ECG. Ultimately, this CNN model may be used as a screening tool for those suspected to suffer from OSA.

Hardware Circuits Design and Performance Evaluation of a Soft Lower Limb Exoskeleton Soft lower limb exoskeletons (LLEs) are wearable devices that have good potential in walking rehabilitation and augmentation. While a few studies focused on the structure design and assistance force optimization of the soft LLEs, rarely work has been conducted on the hardware circuits design. The main purpose of this work is to present a new soft LLE for walking efficiency improvement and introduce its hardware circuits design. A soft LLE for hip flexion assistance and a hardware circuits system with scalability were proposed. To assess the efficacy of the soft LLE, the experimental tests that evaluate the sensor data acquisition, force tracking performance, lower limb muscle activity and metabolic cost were conducted. The time error in the peak assistance force was just 1%. The reduction in the normalized root-mean-square EMG of the rectus femoris was 7.1%. The net metabolic cost in exoskeleton on condition was reduced by 7.8% relative to walking with no exoskeleton. The results show that the designed hardware circuits can be applied to the soft LLE and the soft LLE is able to improve walking efficiency of wearers.

Auc2Charge: An Online Auction Framework for Eectric Vehicle Park-and-Charge The increasing market share of electric vehicles (EVs) makes large-scale charging stations indispensable infrastructure for integrating EVs into the future smart grid. Thus their operation modes have drawn great attention from researchers. One promising mode called park-and-charge was recently proposed. It allows people to park their EVs at a parking lot, where EVs can get charged during the parking time. This mode has been experimented and demonstrated in small scale. However, the missing of an efficient market mechanism is an important gap preventing its large-scale deployment. Existing pricing policies, e.g., pay-by-use and flat-rate pricing, would jeopardize the efficiency of electricity allocation and the corresponding social welfare in the park-and-charge mode, and thus are inapplicable. To find an efficient mechanism, this paper explores the feasibility and benefits of utilizing auction mechanism in the EV park-and-charge mode. The auction allows EV users to submit and update bids on their charging demand to the charging station, which makes corresponding electricity allocation and pricing decisions. To this end, we propose Auc2Charge, an online auction framework. Auc2Charge is truthful and individual rational. Running in polynomial time, it provides an efficient electricity allocation for EV users with a close-form approximation ratio on system social welfare. Through both theoretical analysis and numerical simulation, we demonstrate the efficacy of Auc2Charge in terms of social welfare and user satisfaction.

Competitive on-line scheduling with level of service Motivated by an application in thinwire visualization, we study an abstract on-line scheduling problem where the size of each requested service can be scaled down by the scheduler. Thus, our problem embodies a notion of "Level of Service" that is increasingly important in multimedia applications. We give two schedulers FirstFit and EndFit based on two simple heuristics, and generalize them into a class of greedy schedulers. We show that both FirstFit and EndFit are 2-competitive, and any greedy scheduler is 3-competitive. These bounds are shown to be tight.

Coordinated Charging of Electric Vehicles for Congestion Prevention in the Distribution Grid Distributed energy resources (DERs), like electric vehicles (EVs), can offer valuable services to power systems, such as enabling renewable energy to the electricity producer and providing ancillary services to the system operator. However, these new DERs may challenge the distribution grid due to insufficient capacity in peak hours. This paper aims to coordinate the valuable services and operation constraints of three actors: the EV owner, the Fleet operator (FO) and the Distribution system operator (DSO), considering the individual EV owner's driving requirement, the charging cost of EV and thermal limits of cables and transformers in the proposed market framework. Firstly, a theoretical market framework is described. Within this framework, FOs who represent their customer's (EV owners) interests will centrally guarantee the EV owners' driving requirements and procure the energy for their vehicles with lower cost. The congestion problem will be solved by a coordination between DSO and FOs through a distribution grid capacity market scheme. Then, a mathematical formulation of the market scheme is presented. Further, some case studies are shown to illustrate the effectiveness of the proposed solutions.

Real-Time PEV Charging/Discharging Coordination in Smart Distribution Systems. This paper proposes a novel online coordination method for the charging of plug-in electric vehicles (PEVs) in smart distribution networks. The goal of the proposed method is to optimally charge the PEVs in order to maximize the PEV owners&#39; satisfaction and to minimize system operating costs without violating power system constraints. Unlike the solutions reported in the literature, the proposed c...

Decentralized Plug-in Electric Vehicle Charging Selection Algorithm in Power Systems This paper uses a charging selection concept for plug-in electric vehicles (PEVs) to maximize user convenience levels while meeting predefined circuit-level demand limits. The optimal PEV-charging selection problem requires an exhaustive search for all possible combinations of PEVs in a power system, which cannot be solved for the practical number of PEVs. Inspired by the efficiency of the convex relaxation optimization tool in finding close-to-optimal results in huge search spaces, this paper proposes the application of the convex relaxation optimization method to solve the PEV-charging selection problem. Compared with the results of the uncontrolled case, the simulated results indicate that the proposed PEV-charging selection algorithm only slightly reduces user convenience levels, but significantly mitigates the impact of the PEV-charging on the power system. We also develop a distributed optimization algorithm to solve the PEV-charging selection problem in a decentralized manner, i.e., the binary charging decisions (charged or not charged) are made locally by each vehicle. Using the proposed distributed optimization algorithm, each vehicle is only required to report its power demand rather than report several of its private user state information, mitigating the security problems inherent in such problem. The proposed decentralized algorithm only requires low-speed communication capability, making it suitable for real-time implementation.

Online Coordinated Charging Decision Algorithm for Electric Vehicles Without Future Information The large-scale integration of plug-in electric vehicles (PEVs) to the power grid spurs the need for efficient charging coordination mechanisms. It can be shown that the optimal charging schedule smooths out the energy consumption over time so as to minimize the total energy cost. In practice, however, it is hard to smooth out the energy consumption perfectly, because the future PEV charging demand is unknown at the moment when the charging rate of an existing PEV needs to be determined. In this paper, we propose an online coordinated charging decision (ORCHARD) algorithm, which minimizes the energy cost without knowing the future information. Through rigorous proof, we show that ORCHARD is strictly feasible in the sense that it guarantees to fulfill all charging demands before due time. Meanwhile, it achieves the best known competitive ratio of 2.39. By exploiting the problem structure, we propose a novel reduced-complexity algorithm to replace the standard convex optimization techniques used in ORCHARD. Through extensive simulations, we show that the average performance gap between ORCHARD and the offline optimal solution, which utilizes the complete future information, is as small as 6.5%. By setting a proper speeding factor, the average performance gap can be further reduced to 5%.

Random Forests Random forests are a combination of tree predictors such that each tree depends on the values of a random vector sampled independently and with the same distribution for all trees in the forest. The generalization error for forests converges a.s. to a limit as the number of trees in the forest becomes large. The generalization error of a forest of tree classifiers depends on the strength of the individual trees in the forest and the correlation between them. Using a random selection of features to split each node yields error rates that compare favorably to Adaboost (Y. Freund & R. Schapire, Machine Learning: Proceedings of the Thirteenth International conference, &ast;&ast;&ast;, 148–156), but are more robust with respect to noise. Internal estimates monitor error, strength, and correlation and these are used to show the response to increasing the number of features used in the splitting. Internal estimates are also used to measure variable importance. These ideas are also applicable to regression.

A survey on sensor networks The advancement in wireless communications and electronics has enabled the development of low-cost sensor networks. The sensor networks can be used for various application areas (e.g., health, military, home). For different application areas, there are different technical issues that researchers are currently resolving. The current state of the art of sensor networks is captured in this article, where solutions are discussed under their related protocol stack layer sections. This article also points out the open research issues and intends to spark new interests and developments in this field.

Toward Integrating Vehicular Clouds with IoT for Smart City Services Vehicular ad hoc networks, cloud computing, and the Internet of Things are among the emerging technology enablers offering a wide array of new application possibilities in smart urban spaces. These applications consist of smart building automation systems, healthcare monitoring systems, and intelligent and connected transportation, among others. The integration of IoT-based vehicular technologies will enrich services that are eventually going to ignite the proliferation of exciting and even more advanced technological marvels. However, depending on different requirements and design models for networking and architecture, such integration needs the development of newer communication architectures and frameworks. This work proposes a novel framework for architectural and communication design to effectively integrate vehicular networking clouds with IoT, referred to as VCoT, to materialize new applications that provision various IoT services through vehicular clouds. In this article, we particularly put emphasis on smart city applications deployed, operated, and controlled through LoRaWAN-based vehicular networks. LoraWAN, being a new technology, provides efficient and long-range communication possibilities. The article also discusses possible research issues in such an integration including data aggregation, security, privacy, data quality, and network coverage. These issues must be addressed in order to realize the VCoT paradigm deployment, and to provide insights for investors and key stakeholders in VCoT service provisioning. The article presents deep insights for different real-world application scenarios (i.e., smart homes, intelligent traffic light, and smart city) using VCoT for general control and automation along with their associated challenges. It also presents initial insights, through preliminary results, regarding data and resource management in IoT-based resource constrained environments through vehicular clouds.

Space-time modeling of traffic flow. This paper discusses the application of space-time autoregressive integrated moving average (STARIMA) methodology for representing traffic flow patterns. Traffic flow data are in the form of spatial time series and are collected at specific locations at constant intervals of time. Important spatial characteristics of the space-time process are incorporated in the STARIMA model through the use of weighting matrices estimated on the basis of the distances among the various locations where data are collected. These matrices distinguish the space-time approach from the vector autoregressive moving average (VARMA) methodology and enable the model builders to control the number of the parameters that have to be estimated. The proposed models can be used for short-term forecasting of space-time stationary traffic-flow processes and for assessing the impact of traffic-flow changes on other parts of the network. The three-stage iterative space-time model building procedure is illustrated using 7.5min average traffic flow data for a set of 25 loop-detectors located at roads that direct to the centre of the city of Athens, Greece. Data for two months with different traffic-flow characteristics are modelled in order to determine the stability of the parameter estimation.

A novel full structure optimization algorithm for radial basis probabilistic neural networks. In this paper, a novel full structure optimization algorithm for radial basis probabilistic neural networks (RBPNN) is proposed. Firstly, a minimum volume covering hyperspheres (MVCH) algorithm is proposed to heuristically select the initial hidden layer centers of the RBPNN, and then the recursive orthogonal least square (ROLS) algorithm combined with the particle swarm optimization (PSO) algorithm is adopted to further optimize the initial structure of the RBPNN. Finally, the effectiveness and efficiency of our proposed algorithm are evaluated through a plant species identification task involving 50 plant species.

Completely Pinpointing the Missing RFID Tags in a Time-Efficient Way Radio Frequency Identification (RFID) technology has been widely used in inventory management in many scenarios, e.g., warehouses, retail stores, hospitals, etc. This paper investigates a challenging problem of complete identification of missing tags in large-scale RFID systems. Although this problem has attracted extensive attention from academy and industry, the existing work can hardly satisfy the stringent real-time requirements. In this paper, a Slot Filter-based Missing Tag Identification (SFMTI) protocol is proposed to reconcile some expected collision slots into singleton slots and filter out the expected empty slots as well as the unreconcilable collision slots, thereby achieving the improved time-efficiency. The theoretical analysis is conducted to minimize the execution time of the proposed SFMTI. We then propose a cost-effective method to extend SFMTI to the multi-reader scenarios. The extensive simulation experiments and performance results demonstrate that the proposed SFMTI protocol outperforms the most promising Iterative ID-free Protocol (IIP) by reducing nearly 45% of the required execution time, and is just within a factor of 1.18 from the lower bound of the minimum execution time.

An efficient scheduling scheme for mobile charger in on-demand wireless rechargeable sensor networks. Existing studies on wireless sensor networks (WSNs) have revealed that the limited battery capacity of sensor nodes (SNs) hinders their perpetual operation. Recent findings in the domain of wireless energy transfer (WET) have attracted a lot of attention of academia and industry to cater the lack of energy in the WSNs. The main idea of WET is to restore the energy of SNs using one or more wireless mobile chargers (MCs), which leads to a new paradigm of wireless rechargeable sensor networks (WRSNs). The determination of an optimal order of charging the SNs (i.e., charging schedule) in an on-demand WRSN is a well-known NP-hard problem. Moreover, care must be taken while designing the charging schedule of an MC as requesting SNs introduce both spatial and temporal constraints. In this paper, we first present a Linear Programming (LP) formulation for the problem of scheduling an MC and then propose an efficient solution based on gravitational search algorithm (GSA). Our method is presented with a novel agent representation scheme and an efficient fitness function. We perform extensive simulations on the proposed scheme to demonstrate its effectiveness over two state-of-the-art algorithms, namely first come first serve (FCFS) and nearest job next with preemption (NJNP). The simulation results reveal that the proposed scheme outperforms both the existing algorithms in terms of charging latency. The virtue of our scheme is also proved by the well-known statistical test, analysis of variance (ANOVA), followed by post hoc analysis.

Social Robots for (Second) Language Learning in (Migrant) Primary School Children Especially these days, innovation and support from technology to relieve pressure in education is highly urgent. This study tested the potential advantage of a social robot over a tablet in (second) language learning on performance, engagement, and enjoyment. Shortages in primary education call for new technology solutions. Previous studies combined robots with tablets, to compensate for robot’s limitations, however, this study applied direct human–robot interaction. Primary school children (N = 63, aged 4–6) participated in a 3-wave field experiment with story-telling exercises, either with a semi-autonomous robot (without tablet, using WOz) or a tablet. Results showed increased learning gains over time when training with a social robot, compared to the tablet. Children who trained with a robot were more engaged in the story-telling task and enjoyed it more. Robot’s behavioral style (social or neutral) hardly differed overall, however, seems to vary for high versus low educational abilities. While social robots need sophistication before being implemented in schools, our study shows the potential of social robots as tutors in (second) language learning.

Speed and Accuracy Tradeoff for LiDAR Data Based Road Boundary Detection Road boundary detection is essential for autonomous vehicle localization and decision-making, especially under GPS signal loss and lane discontinuities. For road boundary detection in structural environments, obstacle occlusions and large road curvature are two significant challenges. However, an effective and fast solution for these problems has remained elusive. To solve these problems, a speed ...

Modular Lightweight Network for Road Object Detection Using a Feature Fusion Approach This article presents a modular lightweight network model for road objects detection, such as car, pedestrian, and cyclist, especially when they are far away from the camera and their sizes are small. Great advances have been made for the deep networks, but small objects detection is still a challenging task. In order to solve this problem, a majority of existing methods utilize complicated network or bigger image size, which generally leads to higher computation cost. The proposed network model is referred to as modular feature fusion detector (MFFD), using a fast and efficient network architecture for detecting small objects. The contribution lies in the following aspects: 1) two base modules have been designed for efficient computation: a) Front module reduces the information loss from raw input images and b) Tinier module decreases the model size and computation cost, while ensuring the detection accuracy; 2) by stacking the base modules, we design a context features fusion framework for multiscale object detection; and 3) the proposed method is efficient in terms of model size and computation cost, which is applicable for resource-limited devices, such as embedded systems for advanced driver-assistance systems (ADASs). Comparisons with the state-of-the-art on the challenging KITTI dataset reveal the superiority of the proposed method. Especially, 100 ft/s can be achieved on the embedded GPUs such as Jetson TX2.

L-Shape Fitting-Based Vehicle Pose Estimation and Tracking Using 3D-LiDAR Detecting and tracking moving vehicles is one of the most fundamental functions of autonomous vehicles driving in complex scenarios, as it forms the foundation of decision making and path planning. In order to estimate the pose information of moving vehicles accurately, 3D-LiDAR is widely used for accurate distance data. This paper proposed a real-time tracking algorithm based on L-Shape fitting. ...

PointRCNN: 3D Object Proposal Generation and Detection From Point Cloud In this paper, we propose PointRCNN for 3D object detection from raw point cloud. The whole framework is composed of two stages: stage-1 for the bottom-up 3D proposal generation and stage-2 for refining proposals in the canonical coordinates to obtain the final detection results. Instead of generating proposals from RGB image or projecting point cloud to bird's view or voxels as previous methods do, our stage-1 sub-network directly generates a small number of high-quality 3D proposals from point cloud in a bottom-up manner via segmenting the point cloud of the whole scene into foreground points and background. The stage-2 sub-network transforms the pooled points of each proposal to canonical coordinates to learn better local spatial features, which is combined with global semantic features of each point learned in stage-1 for accurate box refinement and confidence prediction. Extensive experiments on the 3D detection benchmark of KITTI dataset show that our proposed architecture outperforms state-of-the-art methods with remarkable margins by using only point cloud as input. The code is available at https://github.com/sshaoshuai/PointRCNN.

IoU Loss for 2D/3D Object Detection In the 2D/3D object detection task, Intersection-over-Union (IoU) has been widely employed as an evaluation metric to evaluate the performance of different detectors in the testing stage. However, during the training stage, the common distance loss (e.g, L_1 or L_2) is often adopted as the loss function to minimize the discrepancy between the predicted and ground truth Bounding Box (Bbox). To eliminate the performance gap between training and testing, the IoU loss has been introduced for 2D object detection in [1] and [2]. Unfortunately, all these approaches only work for axis-aligned 2D Boxes, which cannot be applied for more general object detection task with rotated Boxes. To resolve this issue, we investigate the IoU computation for two rotated Boxes first and then implement a unified framework, IoU loss layer for both 2D and 3D object detection tasks. By integrating the implemented IoU loss into several state-of-the-art 3D object detectors, consistent improvements have been achieved for both bird-eye-view 2D detection and point cloud 3D detection on the public KITTI [3] benchmark.

MU-GAN: Facial Attribute Editing based on Multi-attention Mechanism Facial attribute editing has mainly two objectives: 1) translating image from a source domain to a target one, and 2) only changing the facial regions related to a target attribute and preserving the attribute-excluding details. In this work, we propose a multi-attention U-Net-based generative adversarial network (MU-GAN). First, we replace a classic convolutional encoder-decoder with a symmetric ...

Lateral Control in Precision Docking Using RTK-GNSS/INS and LiDAR for Localization This article describes the design of a lateral control method for precision docking using GNSS/INS and LiDAR for localization. The aim of this study is to verify whether an automated driving bus can achieve the requirement of standard deviation of the lateral error (1.25 cm) using onboard sensors without additional infrastructure. To design the controller and decide the conditions of pilot tests on public roads, the vehicle dynamics, steering system, sensor dynamics, and disturbances are modelled. The steering system of the vehicle can be approximated as a second-order system, and the measuring time of the sensors are considered, which cause reduction of damping ratio in lateral dynamics. The numerical analysis revealed that the vehicle speed should be reduced to under 4.2 m/s (15 km/h) before precision docking starts, to make the lateral motion of the vehicle more stable. Experimental results show that the standard deviation of lateral error in precision docking using LiDAR for localization is 1.2 cm, which is slightly under the criterion threshold. The results also verify that the vehicle and sensor model describes the vehicle lateral motion appropriately.

A Theoretical Foundation of Intelligence Testing and Its Application for Intelligent Vehicles Intelligent vehicle testing received quickly increasing attention due to the intermittent accidents of intelligent vehicle prototypes that occurred recently. In this paper, we investigate the theoretical underpinnings of such testing and establish a rigid analyzing framework for general intelligence testing problems by borrowing the ideas of Probably Approximately Correct (PAC) learning. Our focus...

A new approach for dynamic fuzzy logic parameter tuning in Ant Colony Optimization and its application in fuzzy control of a mobile robot Central idea is to avoid or slow down full convergence through the dynamic variation of parameters.Performance of different ACO variants was observed to choose one as the basis to the proposed approach.Convergence fuzzy controller with the objective of maintaining diversity to avoid premature convergence was created. Ant Colony Optimization is a population-based meta-heuristic that exploits a form of past performance memory that is inspired by the foraging behavior of real ants. The behavior of the Ant Colony Optimization algorithm is highly dependent on the values defined for its parameters. Adaptation and parameter control are recurring themes in the field of bio-inspired optimization algorithms. The present paper explores a new fuzzy approach for diversity control in Ant Colony Optimization. The main idea is to avoid or slow down full convergence through the dynamic variation of a particular parameter. The performance of different variants of the Ant Colony Optimization algorithm is analyzed to choose one as the basis to the proposed approach. A convergence fuzzy logic controller with the objective of maintaining diversity at some level to avoid premature convergence is created. Encouraging results on several traveling salesman problem instances and its application to the design of fuzzy controllers, in particular the optimization of membership functions for a unicycle mobile robot trajectory control are presented with the proposed method.

Adaptive Navigation Support Adaptive navigation support is a specific group of technologies that support user navigation in hyperspace, by adapting to the goals, preferences and knowledge of the individual user. These technologies, originally developed in the field of adaptive hypermedia, are becoming increasingly important in several adaptive Web applications, ranging from Web-based adaptive hypermedia to adaptive virtual reality. This chapter provides a brief introduction to adaptive navigation support, reviews major adaptive navigation support technologies and mechanisms, and illustrates these with a range of examples.

Rich Models for Steganalysis of Digital Images We describe a novel general strategy for building steganography detectors for digital images. The process starts with assembling a rich model of the noise component as a union of many diverse submodels formed by joint distributions of neighboring samples from quantized image noise residuals obtained using linear and nonlinear high-pass filters. In contrast to previous approaches, we make the model assembly a part of the training process driven by samples drawn from the corresponding cover- and stego-sources. Ensemble classifiers are used to assemble the model as well as the final steganalyzer due to their low computational complexity and ability to efficiently work with high-dimensional feature spaces and large training sets. We demonstrate the proposed framework on three steganographic algorithms designed to hide messages in images represented in the spatial domain: HUGO, edge-adaptive algorithm by Luo , and optimally coded ternary $\\pm {\\hbox{1}}$ embedding. For each algorithm, we apply a simple submodel-selection technique to increase the detection accuracy per model dimensionality and show how the detection saturates with increasing complexity of the rich model. By observing the differences between how different submodels engage in detection, an interesting interplay between the embedding and detection is revealed. Steganalysis built around rich image models combined with ensemble classifiers is a promising direction towards automatizing steganalysis for a wide spectrum of steganographic schemes.

Spatial augmented reality as a method for a mobile robot to communicate intended movement. •Communication strategies are to allow robots to convey upcoming movements to humans.•Arrows for conveying direction of movement are understood by humans.•Simple maps depicting a sequence of upcoming movements are useful to humans.•Robots projecting arrows and a map can effectively communicate upcoming movement.

Bon Appetit! Robot Persuasion for Food Recommendation. The integration of social robots within service industries requires social robots to be persuasive. We conducted a vignette experiment to investigate the persuasiveness of a human, robot, and an information kiosk when offering consumers a restaurant recommendation. We found that embodiment type significantly affects the persuasiveness of the agent, but only when using a specific recommendation sentence. These preliminary results suggest that human-like features of an agent may serve to boost persuasion in recommendation systems. However, the extent of the effect is determined by the nature of the given recommendation.

TD-EUA - Task-Decomposable Edge User Allocation with QoE Optimization.

Enhanced Coordinated Operations of Electric Power and Transportation Networks via EV Charging Services Electric power and transportation networks become increasingly coupled through electric vehicles (EV) charging station (EVCS) as the penetration of EVs continues to grow. In this paper, we propose a holistic framework to enhance the operation of coordinated electric power distribution network (PDN) and urban transportation network (UTN) via EV charging services. Under this framework, a bi-level mo...

Computational difficulties of bilevel linear programming We show, using small examples, that two algorithms previously published for the Bilevel Linear Programming problem BLP may fail to find the optimal solution and thus must be considered to be heuris...

Competitive charging station pricing for plug-in electric vehicles This paper considers the problem of charging station pricing and station selection of plug-in electric vehicles (PEVs). Every PEV needs to select a charging station by considering the charging prices, waiting times, and travel distances. Each charging station optimizes its charging price based on the prediction of the PEVs' charging station selection decisions, in an attempt to maximize its profit. To obtain insights of such a highly coupled system, we consider a one-dimensional system with two charging stations and Poisson arriving PEVs. We propose a multi-leader-multi-follower Stackelberg game model, in which the charging stations (leaders) announce their charging prices in Stage I, and the PEVs (followers) make their charging station selections in Stage II. We show that there always exists a unique charging station selection equilibrium in Stage II, and such equilibrium depends on the price difference between the charging stations. We then characterize the sufficient conditions for the existence and uniqueness of the pricing equilibrium in Stage I. Unfortunately, it is hard to compute the pricing equilibrium in closed form. To overcome this challenge, we develop a low-complexity algorithm that efficiently computes the pricing equilibrium and the subgame perfect equilibrium of our Stackelberg game with no information exchange.

Placement of EV Charging Stations - Balancing Benefits Among Multiple Entities. This paper studies the problem of multistage placement of electric vehicle (EV) charging stations with incremental EV penetration rates. A nested logit model is employed to analyze the charging preference of the individual consumer (EV owner) and predict the aggregated charging demand at the charging stations. The EV charging industry is modeled as an oligopoly where the entire market is dominated...

An Analysis of Price Competition in Heterogeneous Electric Vehicle Charging Stations In this paper, we investigate the price competition among electric vehicle charging stations (EVCSs) with renewable power generators (RPGs). Both a large-sized EVCS (L-EVCS) and small-sized EVCSs (S-EVCSs), which have different capacities, are considered. Moreover, the price elasticity of electric vehicles (EVs), the effect of the distance between an EV and the EVCSs, and the impact of the number ...

PEV Fast-Charging Station Sizing and Placement in Coupled Transportation-Distribution Networks Considering Power Line Conditioning Capability The locations and sizes of plug-in electric vehicle fast-charging stations (PEVFCS) can affect traffic flow in the urban transportation network (TN) and operation indices in the electrical energy distribution network (DN). PEVFCSs are supplied by the DNs, generally using power electronic devices. Thus, PEVFCSs could be used as power line conditioners, especially as active filters (for mitigating harmonic pollutions) and reactive power compensators. Accordingly, this paper proposes a mixed-integer linear programming model taking into account the traffic impacts and power line conditioning capabilities of PEVFCSs to determine optimal locations and sizes of PEVFCSs. Various load profile patterns and the variation of charging demand during the planning horizon are included in this model to consider different operation states of DN and TN. The proposed model is implemented in GAMS and applied to two standard test systems. Numerical results are provided for the case studies and various scenarios. The results confirm the ability and efficiency of the proposed model and its superiority to the existing models.

The Sybil Attack Large-scale peer-to-peer systems facesecurity threats from faulty or hostile remotecomputing elements. To resist these threats, manysuch systems employ redundancy. However, if asingle faulty entity can present multiple identities,it can control a substantial fraction of the system,thereby undermining this redundancy. Oneapproach to preventing these &quot;Sybil attacks&quot; is tohave a trusted agency certify identities. Thispaper shows that, without a logically centralizedauthority, Sybil...

Long short-term memory. Learning to store information over extended time intervals by recurrent backpropagation takes a very long time, mostly because of insufficient, decaying error backflow. We briefly review Hochreiter's (1991) analysis of this problem, then address it by introducing a novel, efficient, gradient-based method called long short-term memory (LSTM). Truncating the gradient where this does not do harm, LSTM can learn to bridge minimal time lags in excess of 1000 discrete-time steps by enforcing constant error flow through constant error carousels within special units. Multiplicative gate units learn to open and close access to the constant error flow. LSTM is local in space and time; its computational complexity per time step and weight is O(1). Our experiments with artificial data involve local, distributed, real-valued, and noisy pattern representations. In comparisons with real-time recurrent learning, back propagation through time, recurrent cascade correlation, Elman nets, and neural sequence chunking, LSTM leads to many more successful runs, and learns much faster. LSTM also solves complex, artificial long-time-lag tasks that have never been solved by previous recurrent network algorithms.

Toward Integrating Vehicular Clouds with IoT for Smart City Services Vehicular ad hoc networks, cloud computing, and the Internet of Things are among the emerging technology enablers offering a wide array of new application possibilities in smart urban spaces. These applications consist of smart building automation systems, healthcare monitoring systems, and intelligent and connected transportation, among others. The integration of IoT-based vehicular technologies will enrich services that are eventually going to ignite the proliferation of exciting and even more advanced technological marvels. However, depending on different requirements and design models for networking and architecture, such integration needs the development of newer communication architectures and frameworks. This work proposes a novel framework for architectural and communication design to effectively integrate vehicular networking clouds with IoT, referred to as VCoT, to materialize new applications that provision various IoT services through vehicular clouds. In this article, we particularly put emphasis on smart city applications deployed, operated, and controlled through LoRaWAN-based vehicular networks. LoraWAN, being a new technology, provides efficient and long-range communication possibilities. The article also discusses possible research issues in such an integration including data aggregation, security, privacy, data quality, and network coverage. These issues must be addressed in order to realize the VCoT paradigm deployment, and to provide insights for investors and key stakeholders in VCoT service provisioning. The article presents deep insights for different real-world application scenarios (i.e., smart homes, intelligent traffic light, and smart city) using VCoT for general control and automation along with their associated challenges. It also presents initial insights, through preliminary results, regarding data and resource management in IoT-based resource constrained environments through vehicular clouds.

Space-time modeling of traffic flow. This paper discusses the application of space-time autoregressive integrated moving average (STARIMA) methodology for representing traffic flow patterns. Traffic flow data are in the form of spatial time series and are collected at specific locations at constant intervals of time. Important spatial characteristics of the space-time process are incorporated in the STARIMA model through the use of weighting matrices estimated on the basis of the distances among the various locations where data are collected. These matrices distinguish the space-time approach from the vector autoregressive moving average (VARMA) methodology and enable the model builders to control the number of the parameters that have to be estimated. The proposed models can be used for short-term forecasting of space-time stationary traffic-flow processes and for assessing the impact of traffic-flow changes on other parts of the network. The three-stage iterative space-time model building procedure is illustrated using 7.5min average traffic flow data for a set of 25 loop-detectors located at roads that direct to the centre of the city of Athens, Greece. Data for two months with different traffic-flow characteristics are modelled in order to determine the stability of the parameter estimation.

Dynamic transfer among alternative controllers and its relation to antiwindup controller design Advanced control strategies and modern consulting provide new challenges for the classical problem of bumpless transfer. It can, for example, be necessary to transfer between an only approximately known existing analog controller and a new digital or adaptive controller without accessing any states. Transfer ought to be bidirectional and not presuppose steady state, so that an immediate back-transfer is possible if the new controller should drive the plant unstable. We present a scheme that meets these requirements. By casting the problem of bidirectional transfer into an associated tracking control problem, systematic analysis and design procedures from control theory can be applied. The associated control problem also has a correspondence to the design of antiwindup controllers. The paper includes laboratory and industrial applications.

A Framework of Joint Mobile Energy Replenishment and Data Gathering in Wireless Rechargeable Sensor Networks Recent years have witnessed the rapid development and proliferation of techniques on improving energy efficiency for wireless sensor networks. Although these techniques can relieve the energy constraint on wireless sensors to some extent, the lifetime of wireless sensor networks is still limited by sensor batteries. Recent studies have shown that energy rechargeable sensors have the potential to provide perpetual network operations by capturing renewable energy from external environments. However, the low output of energy capturing devices can only provide intermittent recharging opportunities to support low-rate data services due to spatial-temporal, geographical or environmental factors. To provide steady and high recharging rates and achieve energy efficient data gathering from sensors, in this paper, we propose to utilize mobility for joint energy replenishment and data gathering. In particular, a multi-functional mobile entity, called SenCarin this paper, is employed, which serves not only as a mobile data collector that roams over the field to gather data via short-range communication but also as an energy transporter that charges static sensors on its migration tour via wireless energy transmissions. Taking advantages of SenCar's controlled mobility, we focus on the joint optimization of effective energy charging and high-performance data collections. We first study this problem in general networks with random topologies. We give a two-step approach for the joint design. In the first step, the locations of a subset of sensors are periodically selected as anchor points, where the SenCar will sequentially visit to charge the sensors at these locations and gather data from nearby sensors in a multi-hop fashion. To achieve a desirable balance between energy replenishment amount and data gathering latency, we provide a selection algorithm to search for a maximum number of anchor points where sensors hold the least battery energy, and meanwhile by visiting them, - he tour length of the SenCar is no more than a threshold. In the second step, we consider data gathering performance when the SenCar migrates among these anchor points. We formulate the problem into a network utility maximization problem and propose a distributed algorithm to adjust data rates at which sensors send buffered data to the SenCar, link scheduling and flow routing so as to adapt to the up-to-date energy replenishing status of sensors. Besides general networks, we also study a special scenario where sensors are regularly deployed. For this case we can provide a simplified solution of lower complexity by exploiting the symmetry of the topology. Finally, we validate the effectiveness of our approaches by extensive numerical results, which show that our solutions can achieve perpetual network operations and provide high network utility.

Finite-Time Adaptive Fuzzy Tracking Control Design for Nonlinear Systems. This paper addresses the finite-time tracking problem of nonlinear pure-feedback systems. Unlike the literature on traditional finite-time stabilization, in this paper the nonlinear system functions, including the bounding functions, are all totally unknown. Fuzzy logic systems are used to model those unknown functions. To present a finite-time control strategy, a criterion of semiglobal practical...

Intention-detection strategies for upper limb exosuits: model-based myoelectric vs dynamic-based control The cognitive human-robot interaction between an exosuit and its wearer plays a key role in determining both the biomechanical effects of the device on movements and its perceived effectiveness. There is a lack of evidence, however, on the comparative performance of different control methods, implemented on the same device. Here, we compare two different control approaches on the same robotic suit: a model-based myoelectric control (myoprocessor), which estimates the joint torque from the activation of target muscles, and a dynamic-based control that provides support against gravity using an inverse dynamic model. Tested on a cohort of four healthy participants, assistance from the exosuit results in a marked reduction in the effort of muscles working against gravity with both control approaches (peak reduction of 68.6±18.8%, for the dynamic arm model and 62.4±25.1% for the myoprocessor), when compared to an unpowered condition. Neither of the two controllers had an affect on the performance of their users in a joint-angle tracking task (peak errors of 15.4° and 16.4° for the dynamic arm model and myoprocessor, respectively, compared to 13.1o in the unpowered condition). However, our results highlight the remarkable adaptability of the myoprocessor to seamlessly adapt to changing external dynamics.

Realization of learning induced self-adaptive sampling in noisy optimization. •Proposed three extensions to amend artificial bee colony in noisy fitness landscape.•Stochastic learning automata induced adaptive sample size selection of trial solution.•Local noise distribution captured by local neighborhood fitness variance.•Effective fitness estimate of solution from local distribution of noisy samples.•Probabilistic selection to avoid deterministic dismissal of quality solutions.

Mobile cloud computing: A survey Despite increasing usage of mobile computing, exploiting its full potential is difficult due to its inherent problems such as resource scarcity, frequent disconnections, and mobility. Mobile cloud computing can address these problems by executing mobile applications on resource providers external to the mobile device. In this paper, we provide an extensive survey of mobile cloud computing research, while highlighting the specific concerns in mobile cloud computing. We present a taxonomy based on the key issues in this area, and discuss the different approaches taken to tackle these issues. We conclude the paper with a critical analysis of challenges that have not yet been fully met, and highlight directions for future work.

Harmony search algorithm for solving Sudoku Harmony search (HS) algorithm was applied to solving Sudoku puzzle. The HS is an evolutionary algorithm which mimics musicians' behaviors such as random play, memory-based play, and pitch-adjusted play when they perform improvisation. Sudoku puzzles in this study were formulated as an optimization problem with number-uniqueness penalties. HS could successfully solve the optimization problem after 285 function evaluations, taking 9 seconds. Also, sensitivity analysis of HS parameters was performed to obtain a better idea of algorithm parameter values.

DEC: dynamically evolving clustering and its application to structure identification of evolving fuzzy models. Identification of models from input-output data essentially requires estimation of appropriate cluster centers. In this paper, a new online evolving clustering approach for streaming data is proposed. Unlike other approaches that consider either the data density or distance from existing cluster centers, this approach uses cluster weight and distance before generating new clusters. To capture the dynamics of the data stream, the cluster weight is defined in both data and time space in such a way that it decays exponentially with time. It also applies concepts from computational geometry to determine the neighborhood information while forming clusters. A distinction is made between core and noncore clusters to effectively identify the real outliers. The approach efficiently estimates cluster centers upon which evolving Takagi-Sugeno models are developed. The experimental results with developed models show that the proposed approach attains results at par or better than existing approaches and significantly reduces the computational overhead, which makes it suitable for real-time applications.

An Easily Understandable Grey Wolf Optimizer and Its Application to Fuzzy Controller Tuning. This paper proposes an easily understandable Grey Wolf Optimizer (GWO) applied to the optimal tuning of the parameters of Takagi-Sugeno proportional-integral fuzzy controllers (T-S PI-FCs). GWO is employed for solving optimization problems focused on the minimization of discrete-time objective functions defined as the weighted sum of the absolute value of the control error and of the squared output sensitivity function, and the vector variable consists of the tuning parameters of the T-S PI-FCs. Since the sensitivity functions are introduced with respect to the parametric variations of the process, solving these optimization problems is important as it leads to fuzzy control systems with a reduced process parametric sensitivity obtained by a GWO-based fuzzy controller tuning approach. GWO algorithms applied with this regard are formulated in easily understandable terms for both vector and scalar operations, and discussions on stability, convergence, and parameter settings are offered. The controlled processes referred to in the course of this paper belong to a family of nonlinear servo systems, which are modeled by second order dynamics plus a saturation and dead zone static nonlinearity. Experimental results concerning the angular position control of a laboratory servo system are included for validating the proposed method.

Evolutionary Wall-Following Hexapod Robot Using Advanced Multiobjective Continuous Ant Colony Optimized Fuzzy Controller. This paper proposes an evolutionary wall-following hexapod robot, where a new multiobjective evolutionary fuzzy control approach is proposed to control both walking orientation and speed of a hexapod robot for a wall-following task. According to the measurements of four distance sensors, a fuzzy controller (FC) controls the walking speed of the robot by changing the common swing angles of its six ...

Fuzzy Logic in Dynamic Parameter Adaptation of Harmony Search Optimization for Benchmark Functions and Fuzzy Controllers. Nowadays the use of fuzzy logic has been increasing in popularity, and this is mainly due to the inference mechanism that allows simulating human reasoning in knowledge-based systems. The main contribution of this work is using the concepts of fuzzy logic in a method for dynamically adapting the main parameters of the harmony search algorithm during execution. Dynamical adaptation of parameters in metaheuristics has been shown to improve performance and accuracy in a wide range of applications. For this reason, we propose and approach for fuzzy adaptation of parameters in harmony search. Two case studies are considered for testing the proposed approach, the optimization of mathematical functions, which are unimodal, multimodal, hybrid, and composite functions and a control problem without noise and when noise is considered. A statistical comparison between the harmony search algorithm and the fuzzy harmony search algorithm is presented to verify the advantages of the proposed approach.

Finite-Time Input-to-State Stability and Applications to Finite-Time Control Design This paper extends the well-known concept, Sontag's input-to-state stability (ISS), to finite-time control problems. In other words, a new concept, finite-time input-to-state stability (FTISS), is proposed and then is applied to both the analysis of finite-time stability and the design of finite-time stabilizing feedback laws of control systems. With finite-time stability, nonsmoothness has to be considered, and serious technical challenges arise in the design of finite-time controllers and the stability analysis of the closed-loop system. It is found that FTISS plays an important role as the conventional ISS in the context of asymptotic stability analysis and smooth feedback stabilization. Moreover, a robust adaptive controller is proposed to handle nonlinear systems with parametric and dynamic uncertainties by virtue of FTISS and related arguments.

Adam: A Method for Stochastic Optimization. We introduce Adam, an algorithm for first-order gradient-based optimization of stochastic objective functions, based on adaptive estimates of lower-order moments. The method is straightforward to implement, is computationally efficient, has little memory requirements, is invariant to diagonal rescaling of the gradients, and is well suited for problems that are large in terms of data and/or parameters. The method is also appropriate for non-stationary objectives and problems with very noisy and/or sparse gradients. The hyper-parameters have intuitive interpretations and typically require little tuning. Some connections to related algorithms, on which Adam was inspired, are discussed. We also analyze the theoretical convergence properties of the algorithm and provide a regret bound on the convergence rate that is comparable to the best known results under the online convex optimization framework. Empirical results demonstrate that Adam works well in practice and compares favorably to other stochastic optimization methods. Finally, we discuss AdaMax, a variant of Adam based on the infinity norm.

Multiple Lyapunov functions and other analysis tools for switched and hybrid systems In this paper, we introduce some analysis tools for switched and hybrid systems. We first present work on stability analysis. We introduce multiple Lyapunov functions as a tool for analyzing Lyapunov stability and use iterated function systems (IFS) theory as a tool for Lagrange stability. We also discuss the case where the switched systems are indexed by an arbitrary compact set. Finally, we extend Bendixson's theorem to the case of Lipschitz continuous vector fields, allowing limit cycle analysis of a class of "continuous switched" systems.

Learning to Predict Driver Route and Destination Intent For many people, driving is a routine activity where people drive to the same destinations using the same routes on a regular basis. Many drivers, for example, will drive to and from work along a small set of routes, at about the same time every day of the working week. Similarly, although a person may shop on different days or at different times, they will often visit the same grocery store(s). In this paper, we present a novel approach to predicting driver intent that exploits the predictable nature of everyday driving. Our approach predicts a driver's intended route and destination through the use of a probabilistic model learned from observation of their driving habits. We show that by using a low-cost GPS sensor and a map database, it is possible to build a hidden Markov model (HMM) of the routes and destinations used by the driver. Furthermore, we show that this model can be used to make accurate predictions of the driver's destination and route through on-line observation of their GPS position during the trip. We present a thorough evaluation of our approach using a corpus of almost a month of real, everyday driving. Our results demonstrate the effectiveness of the approach, achieving approximately 98% accuracy in most cases. Such high performance suggests that the method can be harnessed for improved safety monitoring, route planning taking into account traffic density, and better trip duration prediction

Software-Defined Networking: A Comprehensive Survey The Internet has led to the creation of a digital society, where (almost) everything is connected and is accessible from anywhere. However, despite their widespread adoption, traditional IP networks are complex and very hard to manage. It is both difficult to configure the network according to predefined policies, and to reconfigure it to respond to faults, load, and changes. To make matters even more difficult, current networks are also vertically integrated: the control and data planes are bundled together. Software-defined networking (SDN) is an emerging paradigm that promises to change this state of affairs, by breaking vertical integration, separating the network's control logic from the underlying routers and switches, promoting (logical) centralization of network control, and introducing the ability to program the network. The separation of concerns, introduced between the definition of network policies, their implementation in switching hardware, and the forwarding of traffic, is key to the desired flexibility: by breaking the network control problem into tractable pieces, SDN makes it easier to create and introduce new abstractions in networking, simplifying network management and facilitating network evolution. In this paper, we present a comprehensive survey on SDN. We start by introducing the motivation for SDN, explain its main concepts and how it differs from traditional networking, its roots, and the standardization activities regarding this novel paradigm. Next, we present the key building blocks of an SDN infrastructure using a bottom-up, layered approach. We provide an in-depth analysis of the hardware infrastructure, southbound and northbound application programming interfaces (APIs), network virtualization layers, network operating systems (SDN controllers), network programming languages, and network applications. We also look at cross-layer problems such as debugging and troubleshooting. In an effort to anticipate the future evolution of this - ew paradigm, we discuss the main ongoing research efforts and challenges of SDN. In particular, we address the design of switches and control platforms - with a focus on aspects such as resiliency, scalability, performance, security, and dependability - as well as new opportunities for carrier transport networks and cloud providers. Last but not least, we analyze the position of SDN as a key enabler of a software-defined environment.

Deep Learning in Mobile and Wireless Networking: A Survey. The rapid uptake of mobile devices and the rising popularity of mobile applications and services pose unprecedented demands on mobile and wireless networking infrastructure. Upcoming 5G systems are evolving to support exploding mobile traffic volumes, real-time extraction of fine-grained analytics, and agile management of network resources, so as to maximize user experience. Fulfilling these tasks is challenging, as mobile environments are increasingly complex, heterogeneous, and evolving. One potential solution is to resort to advanced machine learning techniques, in order to help manage the rise in data volumes and algorithm-driven applications. The recent success of deep learning underpins new and powerful tools that tackle problems in this space. In this paper we bridge the gap between deep learning and mobile and wireless networking research, by presenting a comprehensive survey of the crossovers between the two areas. We first briefly introduce essential background and state-of-the-art in deep learning techniques with potential applications to networking. We then discuss several techniques and platforms that facilitate the efficient deployment of deep learning onto mobile systems. Subsequently, we provide an encyclopedic review of mobile and wireless networking research based on deep learning, which we categorize by different domains. Drawing from our experience, we discuss how to tailor deep learning to mobile environments. We complete this survey by pinpointing current challenges and open future directions for research.

Attitudes Towards Social Robots In Education: Enthusiast, Practical, Troubled, Sceptic, And Mindfully Positive While social robots bring new opportunities for education, they also come with moral challenges. Therefore, there is a need for moral guidelines for the responsible implementation of these robots. When developing such guidelines, it is important to include different stakeholder perspectives. Existing (qualitative) studies regarding these perspectives however mainly focus on single stakeholders. In this exploratory study, we examine and compare the attitudes of multiple stakeholders on the use of social robots in primary education, using a novel questionnaire that covers various aspects of moral issues mentioned in earlier studies. Furthermore, we also group the stakeholders based on similarities in attitudes and examine which socio-demographic characteristics influence these attitude types. Based on the results, we identify five distinct attitude profiles and show that the probability of belonging to a specific profile is affected by such characteristics as stakeholder type, age, education and income. Our results also indicate that social robots have the potential to be implemented in education in a morally responsible way that takes into account the attitudes of various stakeholders, although there are multiple moral issues that need to be addressed first. Finally, we present seven (practical) implications for a responsible application of social robots in education following from our results. These implications provide valuable insights into how social robots should be implemented.

Vehicle Position and Context Detection Using V2V Communication A pre-crash detection and warning system in a host vehicle needs to accurately determine the position of each remote vehicle in its vicinity and the context of the driving environment. ADAS (Advanced driver-assistance systems) have extensively used camera radar and LIDAR for automatic detection of vehicles, pedestrians and other road users and their behaviors. However, these vehicle-resident senso...

Real-time Localization in Outdoor Environments using Stereo Vision and Inexpensive GPS We describe a real-time, low-cost system to localize a mobile robot in outdoor environments. Our system relies on stereo vision to robustly estimate frame-to-frame motion in real time (also known as visual odometry). The motion estimation problem is formulated efficiently in the disparity space and results in accurate and robust estimates of the motion even for a small-baseline configuration. Our system uses inertial measurements to fill in motion estimates when visual odometry fails. This incremental motion is then fused with a low-cost GPS sensor using a Kalman Filter to prevent long-term drifts. Experimental results are presented for outdoor localization in moderately sized environments (\geqslant 100 meters)

Vision based robot localization by ground to satellite matching in GPS-denied situations This paper studies the problem of matching images captured from an unmanned ground vehicle (UGV) to those from a satellite or high-flying vehicle. We focus on situations where the UGV navigates in remote areas with few man-made structures. This is a difficult problem due to the drastic change in perspective between the ground and aerial imagery and the lack of environmental features for image comparison. We do not rely on GPS, which may be jammed or uncertain. We propose a two-step approach: (1) the UGV images are warped to obtain a bird's eye view of the ground, and (2) this view is compared to a grid of satellite locations using whole-image descriptors. We analyze the performance of a variety of descriptors for different satellite map sizes and various terrain and environment types. We incorporate the air-ground matching into a particle-filter framework for localization using the best-performing descriptor. The results show that vision-based UGV localization from satellite maps is not only possible, but often provides better position estimates than GPS estimates, enabling us to improve the location estimates of Google Street View.

Federated Learning in Vehicular Networks: Opportunities and Solutions The emerging advances in personal devices and privacy concerns have given the rise to the concept of Federated Learning. Federated Learning proves its effectiveness and privacy preservation through collaborative local training and updating a shared machine learning model while protecting the individual data-sets. This article investigates a new type of vehicular network concept, namely a Federated Vehicular Network (FVN), which can be viewed as a robust distributed vehicular network. Compared to traditional vehicular networks, an FVN has centralized components and utilizes both DSRC and mmWave communication to achieve more scalable and stable performance. As a result, FVN can be used to support data-/computation-intensive applications such as distributed machine learning and Federated Learning. The article first outlines the enabling technologies of FVN. Then, we briefly discuss the high-level architecture of FVN and explain why such an architecture is adequate for Federated Learning. In addition, we use auxiliary Blockchain-based systems to facilitate transactions and mitigate malicious behaviors. Next, we discuss in detail one key component of FVN, a federated vehicular cloud (FVC), that is used for sharing data and models in FVN. In particular, we focus on the routing inside FVCs and present our solutions and preliminary evaluation results. Finally, we point out open problems and future research directions of this disruptive technology.

Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions Although convolutional neural networks (CNNs) have achieved great success in computer vision, this work investigates a simpler, convolution-free backbone network use-fid for many dense prediction tasks. Unlike the recently-proposed Vision Transformer (ViT) that was designed for image classification specifically, we introduce the Pyramid Vision Transformer (PVT), which overcomes the difficulties of...

Toward Lightweight, Privacy-Preserving Cooperative Object Classification for Connected Autonomous Vehicles Collaborative perception enables autonomous vehicles to exchange sensor data among each other to achieve cooperative object classification, which is considered an effective means to improve the perception accuracy of connected autonomous vehicles (CAVs). To protect information privacy in cooperative perception, we propose a lightweight, privacy-preserving cooperative object classification framework that allows CAVs to exchange raw sensor data (e.g., images captured by HD camera), without leaking private information. Leveraging chaotic encryption and additive secret sharing technique, image data are first encrypted into two ciphertexts and processed, in the encrypted format, by two separate edge servers. The use of chaotic mapping can avoid information leakage during data uploading. The encrypted images are then processed by the proposed privacy-preserving convolutional neural network (P-CNN) model embedded in the designed secure computing protocols. Finally, the processed results are combined/decrypted on the receiving vehicles to realize cooperative object classification. We formally prove the correctness and security of the proposed framework and carry out intensive experiments to evaluate its performance. The experimental results indicate that P-CNN offers exactly almost the same object classification results as the original CNN model, while offering great privacy protection of shared data and lightweight execution efficiency.

Robust Camera Pose Estimation for Unordered Road Scene Images in Varying Viewing Conditions For continuous performance optimization of camera sensor systems in automated driving, training data from rare corner cases occurring in series production cars are required. In this article, we propose collaborative acquisition of camera images via connected car fleets for synthesis of image sequences from arbitrary road sections which are challenging for machine vision. While allowing a scalable hardware architecture inside the cars, this concept demands to reconstruct the recording locations of the individual images aggregated in the back-end. Varying environmental conditions, dynamic scenes, and small numbers of significant landmarks may hamper camera pose estimation through sparse reconstruction from unordered road scene images. Tackling those problems, we extend a state-of-the-art Structure from Motion pipeline by selecting keypoints based on a semantic image segmentation and removing GPS outliers. We present three challenging image datasets recorded on repetitive test drives under differing environmental conditions for evaluation of our method. The results demonstrate that our optimized pipeline is able to reconstruct the camera viewpoints robustly in the majority of road scenes observed while preserving high image registration rates. Reducing the median deviation from GPS measurements by over 48% for car fleet images, the method increases the accuracy of camera poses dramatically.

A standalone RFID Indoor Positioning System Using Passive Tags Indoor positioning systems (IPSs) locate objects in closed structures such as office buildings, hospitals, stores, factories, and warehouses, where Global Positioning System devices generally do not work. Most available systems apply wireless concepts, optical tracking, and/or ultrasound. This paper presents a standalone IPS using radio frequency identification (RFID) technology. The concept is ba...

Dyme: Dynamic Microservice Scheduling in Edge Computing Enabled IoT In recent years, the rapid development of mobile edge computing (MEC) provides an efficient execution platform at the edge for Internet-of-Things (IoT) applications. Nevertheless, the MEC also provides optimal resources to different microservices, however, underlying network conditions and infrastructures inherently affect the execution process in MEC. Therefore, in the presence of varying network conditions, it is necessary to optimally execute the available task of end users while maximizing the energy efficiency in edge platform and we also need to provide fair Quality-of-Service (QoS). On the other hand, it is necessary to schedule the microservices dynamically to minimize the total network delay and network price. Thus, in this article, unlike most of the existing works, we propose a dynamic microservice scheduling scheme for MEC. We design the microservice scheduling framework mathematically and also discuss the computational complexity of the scheduling algorithm. Extensive simulation results show that the microservice scheduling framework significantly improves the performance metrics in terms of total network delay, average price, satisfaction level, energy consumption rate (ECR), failure rate, and network throughput over other existing baselines.

Reciprocal N-body Collision Avoidance In this paper, we present a formal approach to reciprocal n-body collision avoidance, where multiple mobile robots need to avoid collisions with each other while moving in a common workspace. In our formulation, each robot acts fully in- dependently, and does not communicate with other robots. Based on the definition of velocity obstacles (5), we derive sufficient conditions for collision-free motion by reducing the problem to solving a low-dimensional linear program. We test our approach on several dense and complex simulation scenarios involving thousands of robots and compute collision-free actions for all of them in only a few millisec- onds. To the best of our knowledge, this method is the first that can guarantee local collision-free motion for a large number of robots in a cluttered workspace.

RFID-based techniques for human-activity detection The iBracelet and the Wireless Identification and Sensing Platform promise the ability to infer human activity directly from sensor readings.

RECIFE-MILP: An Effective MILP-Based Heuristic for the Real-Time Railway Traffic Management Problem The real-time railway traffic management problem consists of selecting appropriate train routes and schedules for minimizing the propagation of delay in case of traffic perturbation. In this paper, we tackle this problem by introducing RECIFE-MILP, a heuristic algorithm based on a mixed-integer linear programming model. RECIFE-MILP uses a model that extends one we previously proposed by including additional elements characterizing railway reality. In addition, it implements performance boosting methods selected among several ones through an algorithm configuration tool. We present a thorough experimental analysis that shows that the performances of RECIFE-MILP are better than the ones of the currently implemented traffic management strategy. RECIFE-MILP often finds the optimal solution to instances within the short computation time available in real-time applications. Moreover, RECIFE-MILP is robust to its configuration if an appropriate selection of the combination of boosting methods is performed.

A Covert Channel Over VoLTE via Adjusting Silence Periods. Covert channels represent unforeseen communication methods that exploit authorized overt communication as the carrier medium for covert messages. Covert channels can be a secure and effective means of transmitting confidential information hidden in overt traffic. For covert timing channel, the covert message is usually modulated into inter-packet delays (IPDs) of legitimate traffic, which is not suitable for voice over LTE (VoLTE) since the IPDs of VoLTE traffic are fixed to lose the possibility of being modulated. For this reason, we propose a covert channel via adjusting silence periods, which modulates covert message by the postponing or extending silence periods in VoLTE traffic. To keep the robustness, we employ the Gray code to encode the covert message to reduce the impact of packet loss. Moreover, the proposed covert channel enables the tradeoff between the robustness and voice quality which is an important performance indicator for VoLTE. The experiment results show that the proposed covert channel is undetectable by statistical tests and outperforms the other covert channels based on IPDs in terms of robustness.

Pricing-Based Channel Selection for D2D Content Sharing in Dynamic Environment In order to make device-to-device (D2D) content sharing give full play to its advantage of improving local area services, one of the important issues is to decide the channels that D2D pairs occupy. Most existing works study this issue in static environment, and ignore the guidance for D2D pairs to select the channel adaptively. In this paper, we investigate this issue in dynamic environment where...

Densely Semantically Aligned Person Re-Identification We propose a densely semantically aligned person re-identification (re-ID) framework. It fundamentally addresses the body misalignment problem caused by pose/viewpoint variations, imperfect person detection, occlusion, etc.. By leveraging the estimation of the dense semantics of a person image, we construct a set of densely semantically aligned part images (DSAP-images), where the same spatial positions have the same semantics across different person images. We design a two-stream network that consists of a main full image stream (MF-Stream) and a densely semantically-aligned guiding stream (DSAG-Stream). The DSAG-Stream, with the DSAP-images as input, acts as a regulator to guide the MF-Stream to learn densely semantically aligned features from the original image. In the inference, the DSAG-Stream is discarded and only the MF-Stream is needed, which makes the inference system computationally efficient and robust. To our best knowledge, we are the first to make use of fine grained semantics for addressing misalignment problems for re-ID. Our method achieves rank-1 accuracy of 78.9% (new protocol) on the CUHK03 dataset, 90.4% on the CUHK01 dataset, and 95.7% on the Market1501 dataset, outperforming state-of-the-art methods.

Perceive Where to Focus: Learning Visibility-Aware Part-Level Features for Partial Person Re-Identification This paper considers a realistic problem in person re-identification (re-ID) task, i.e., partial re-ID. Under partial re-ID scenario, the images may contain a partial observation of a pedestrian. If we directly compare a partial pedestrian image with a holistic one, the extreme spatial misalignment significantly compromises the discriminative ability of the learned representation. We propose a Visibility-aware Part Model (VPM) for partial re-ID, which learns to perceive the visibility of regions through self-supervision. The visibility awareness allows VPM to extract region-level features and compare two images with focus on their shared regions (which are visible on both images). VPM gains two-fold benefit toward higher accuracy for partial re-ID. On the one hand, compared with learning a global feature, VPM learns region-level features and thus benefits from fine-grained information. On the other hand, with visibility awareness, VPM is capable to estimate the shared regions between two images and thus suppresses the spatial misalignment. Experimental results confirm that our method significantly improves the learned feature representation and the achieved accuracy is on par with the state of the art.

High-Order Information Matters: Learning Relation And Topology For Occluded Person Re-Identification Occluded person re-identification (ReID) aims to match occluded person images to holistic ones across disjoint cameras. In this paper, we propose a novel framework by learning high-order relation and topology information for discriminative features and robust alignment. At first, we use a CNN backbone and a key-points estimation model to extract semantic local features. Even so, occluded images still suffer from occlusion and outliers. Then, we view the local features of an image as nodes of a graph and propose an adaptive direction graph convolutional (ADGC) layer to pass relation information between nodes. The proposed ADGC layer can automatically suppress the message passing of meaningless features by dynamically learning direction and degree of linkage. When aligning two groups of local features from two images, we view it as a graph matching problem and propose a cross-graph embedded-alignment (CGEA) layer to jointly learn and embed topology information to local features, and straightly predict similarity score. The proposed CGEA layer not only take full use of alignment learned by graph matching but also replace sensitive one-to-one matching with a robust soft one. Finally, extensive experiments on occluded, partial, and holistic ReID tasks show the effectiveness of our proposed method. Specifically, our framework significantly outperforms state-of-the-art by 6.5% mAP scores on Occluded-Duke dataset. Code is available at https://github.com/wangguanan/HOReID.

Omni-Scale Feature Learning For Person Re-Identification As an instance-level recognition problem, person reidentification (ReID) relies on discriminative features, which not only capture different spatial scales but also encapsulate an arbitrary combination of multiple scales. We callse features of both homogeneous and heterogeneous scales omni-scale features. In this paper, a novel deep ReID CNN is designed, termed Omni-Scale Network (OSNet), for omni-scale feature learning. This is achieved by designing a residual block composed of multiple convolutional feature streams, each detecting features at a certain scale. Importantly, a novel unified aggregation gate is introduced to dynamically fuse multiscale features with input-dependent channel-wise weights. To efficiently learn spatial-channel correlations and avoid overfitting, the building block uses both pointwise and depthwise convolutions. By stacking such blocks layer-by-layer, our OSNet is extremely lightweight and can be trained from scratch on existing ReID benchmarks. Despite its small model size, our OSNet achieves state-of-the-art performance on six person-ReID datasets. Code and models are available at: https://github.com/ KaiyangZhou/deep-person-reid.

A Novel Unsupervised Camera-aware Domain Adaptation Framework for Person Re-identification. Unsupervised cross-domain person re-identification (Re-ID) faces two key issues. One is the data distribution discrepancy between source and target domains, and the other is the lack of labelling information in target domain. They are addressed in this paper from the perspective of representation learning. For the first issue, we highlight the presence of camera-level sub-domains as a unique characteristic of person Re-ID, and develop camera-aware domain adaptation to reduce the discrepancy not only between source and target domains but also across these sub-domains. For the second issue, we exploit the temporal continuity in each camera of target domain to create discriminative information. This is implemented by dynamically generating online triplets within each batch, in order to maximally take advantage of the steadily improved feature representation in training process. Together, the above two methods give rise to a novel unsupervised deep domain adaptation framework for person Re-ID. Experiments and ablation studies on benchmark datasets demonstrate its superiority and interesting properties.

Diverse Part Discovery: Occluded Person Re-identification with Part-Aware Transformer Occluded person re-identification (Re-ID) is a challenging task as persons are frequently occluded by various obstacles or other persons, especially in the crowd scenario. To address these issues, we propose a novel end-to-end Part-Aware Transformer (PAT) for occluded person Re-ID through diverse part discovery via a transformer encoder-decoder architecture, including a pixel context based transformer encoder and a part prototype based transformer decoder. The proposed PAT model enjoys several merits. First, to the best of our knowledge, this is the first work to exploit the transformer encoder-decoder architecture for occluded person Re-ID in a unified deep model. Second, to learn part prototypes well with only identity labels, we design two effective mechanisms including part diversity and part discriminability. Consequently, we can achieve diverse part discovery for occluded person Re-ID in a weakly supervised manner. Extensive experimental results on six challenging benchmarks for three tasks (occluded, partial and holistic Re-ID) demonstrate that our proposed PAT performs favorably against stat-of-the-art methods.

Learning Discriminative Features with Multiple Granularities for Person Re-Identification. The combination of global and partial features has been an essential solution to improve discriminative performances in person re-identification (Re-ID) tasks. Previous part-based methods mainly focus on locating regions with specific pre-defined semantics to learn local representations, which increases learning difficulty but not efficient or robust to scenarios with large variances. In this paper, we propose an end-to-end feature learning strategy integrating discriminative information with various granularities. We carefully design the Multiple Granularity Network (MGN), a multi-branch deep network architecture consisting of one branch for global feature representations and two branches for local feature representations. Instead of learning on semantic regions, we uniformly partition the images into several stripes, and vary the number of parts in different local branches to obtain local feature representations with multiple granularities. Comprehensive experiments implemented on the mainstream evaluation datasets including Market-1501, DukeMTMC-reid and CUHK03 indicate that our method robustly achieves state-of-the-art performances and outperforms any existing approaches by a large margin. For example, on Market-1501 dataset in single query mode, we obtain a top result of Rank-1/mAP=96.6%/94.2% with this method after re-ranking.

Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation Object detection performance, as measured on the canonical PASCAL VOC dataset, has plateaued in the last few years. The best-performing methods are complex ensemble systems that typically combine multiple low-level image features with high-level context. In this paper, we propose a simple and scalable detection algorithm that improves mean average precision (mAP) by more than 30% relative to the previous best result on VOC 2012 -- achieving a mAP of 53.3%. Our approach combines two key insights: (1) one can apply high-capacity convolutional neural networks (CNNs) to bottom-up region proposals in order to localize and segment objects and (2) when labeled training data is scarce, supervised pre-training for an auxiliary task, followed by domain-specific fine-tuning, yields a significant performance boost. Since we combine region proposals with CNNs, we call our method R-CNN: Regions with CNN features. We also present experiments that provide insight into what the network learns, revealing a rich hierarchy of image features. Source code for the complete system is available at http://www.cs.berkeley.edu/~rbg/rcnn.

Fog computing and its role in the internet of things Fog Computing extends the Cloud Computing paradigm to the edge of the network, thus enabling a new breed of applications and services. Defining characteristics of the Fog are: a) Low latency and location awareness; b) Wide-spread geographical distribution; c) Mobility; d) Very large number of nodes, e) Predominant role of wireless access, f) Strong presence of streaming and real time applications, g) Heterogeneity. In this paper we argue that the above characteristics make the Fog the appropriate platform for a number of critical Internet of Things (IoT) services and applications, namely, Connected Vehicle, Smart Grid, Smart Cities, and, in general, Wireless Sensors and Actuators Networks (WSANs).

EDUCO - A Collaborative Learning Environment Based on Social Navigation Web-based learning is primarily a lonesome activity, even when it involves working in groups. This is due to the fact that the majority of web-based learning relies on asynchronous forms of interacting with other people. In most of the cases, the chat discussion is the only form of synchronous interaction that adds to the feeling that there are other people present in the environment. EDUCO is a system that tries to bring in the sense of other users in a collaborative learning environment by making the other users and their the navigation visible to everyone else in the environment in real-time. The paper describes EDUCO and presents the first empirical evaluation as EDUCO was used in a university course.

A dual neural network for redundancy resolution of kinematically redundant manipulators subject to joint limits and joint velocity limits. In this paper, a recurrent neural network called the dual neural network is proposed for online redundancy resolution of kinematically redundant manipulators. Physical constraints such as joint limits and joint velocity limits, together with the drift-free criterion as a secondary task, are incorporated into the problem formulation of redundancy resolution. Compared to other recurrent neural networks, the dual neural network is piecewise linear and has much simpler architecture with only one layer of neurons. The dual neural network is shown to be globally (exponentially) convergent to optimal solutions. The dual neural network is simulated to control the PA10 robot manipulator with effectiveness demonstrated.

FreeTagpaper: a pen-and-paper-based collaboration system using visual tags printed on paper We propose a novel collaboration system using pen and paper. FreeTagpaper uses a camera set above a work table to recognize the visual tags printed on the paper to acquire the paper's ID and posture. From the paper ID, FreeTagpaper can retrieve the paper's size and format from a document management database. It then uses the format information to transmit images of handwritten notes to the other side and simultaneously projects the other side's handwritten notes onto this paper. Based on this mutual image sharing scheme, users can collaborate in real time using paper of any size, format, and posture. This report describes preliminary work on FreeTagpaper including the current implementation and essential experimental results.

Unmanned Aerial Vehicle-Aided Communications: Joint Transmit Power and Trajectory Optimization. This letter investigates the transmit power and trajectory optimization problem for unmanned aerial vehicle (UAV)-aided networks. Different from majority of the existing studies with fixed communication infrastructure, a dynamic scenario is considered where a flying UAV provides wireless services for multiple ground nodes simultaneously. To fully exploit the controllable channel variations provide...

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

Evaluating Heuristics for Grid Workflow Scheduling Job scheduling plays a critical role in utilising resources in grid computing environments. However, the heterogeneity of grid resources adds some challenges to the work of job scheduling especially when jobs have dependencies. It is widely recognised that scheduling m jobs to n resources with an objective to achieve a minimum makespan has shown to be NP-complete requiring the development of heuristics. Although a number of heuristics are available for job scheduling optimisation, selecting the best heuristic to use in a given grid environment remains a difficult problem due to the fact that the performance of each original heuristic is usually evaluated under different assumptions. This paper evaluates 12 representative heuristics for dependent job scheduling under one set of common assumptions. The results are presented and analysed which provides an even basis in comparison of the performance of those heuristics.

The Sybil Attack Large-scale peer-to-peer systems facesecurity threats from faulty or hostile remotecomputing elements. To resist these threats, manysuch systems employ redundancy. However, if asingle faulty entity can present multiple identities,it can control a substantial fraction of the system,thereby undermining this redundancy. Oneapproach to preventing these &quot;Sybil attacks&quot; is tohave a trusted agency certify identities. Thispaper shows that, without a logically centralizedauthority, Sybil...

BLEU: a method for automatic evaluation of machine translation Human evaluations of machine translation are extensive but expensive. Human evaluations can take months to finish and involve human labor that can not be reused. We propose a method of automatic machine translation evaluation that is quick, inexpensive, and language-independent, that correlates highly with human evaluation, and that has little marginal cost per run. We present this method as an automated understudy to skilled human judges which substitutes for them when there is need for quick or frequent evaluations.

Computational thinking Summary form only given. My vision for the 21st century, Computational Thinking, will be a fundamental skill used by everyone in the world. To reading, writing, and arithmetic, we should add computational thinking to every child's analytical ability. Computational thinking involves solving problems, designing systems, and understanding human behavior by drawing on the concepts fundamental to computer science. Thinking like a computer scientist means more than being able to program a computer. It requires the ability to abstract and thus to think at multiple levels of abstraction. In this talk I will give many examples of computational thinking, argue that it has already influenced other disciplines, and promote the idea that teaching computational thinking can not only inspire future generations to enter the field of computer science but benefit people in all fields.

Fuzzy logic in control systems: fuzzy logic controller. I.

Switching between stabilizing controllers This paper deals with the problem of switching between several linear time-invariant (LTI) controllers—all of them capable of stabilizing a speci4c LTI process—in such a way that the stability of the closed-loop system is guaranteed for any switching sequence. We show that it is possible to 4nd realizations for any given family of controller transfer matrices so that the closed-loop system remains stable, no matter how we switch among the controller. The motivation for this problem is the control of complex systems where con8icting requirements make a single LTI controller unsuitable. ? 2002 Published by Elsevier Science Ltd.

Tabu Search - Part I

Bidirectional recurrent neural networks In the first part of this paper, a regular recurrent neural network (RNN) is extended to a bidirectional recurrent neural network (BRNN). The BRNN can be trained without the limitation of using input information just up to a preset future frame. This is accomplished by training it simultaneously in positive and negative time direction. Structure and training procedure of the proposed network are explained. In regression and classification experiments on artificial data, the proposed structure gives better results than other approaches. For real data, classification experiments for phonemes from the TIMIT database show the same tendency. In the second part of this paper, it is shown how the proposed bidirectional structure can be easily modified to allow efficient estimation of the conditional posterior probability of complete symbol sequences without making any explicit assumption about the shape of the distribution. For this part, experiments on real data are reported

An intensive survey of fair non-repudiation protocols With the phenomenal growth of the Internet and open networks in general, security services, such as non-repudiation, become crucial to many applications. Non-repudiation services must ensure that when Alice sends some information to Bob over a network, neither Alice nor Bob can deny having participated in a part or the whole of this communication. Therefore a fair non-repudiation protocol has to generate non-repudiation of origin evidences intended to Bob, and non-repudiation of receipt evidences destined to Alice. In this paper, we clearly define the properties a fair non-repudiation protocol must respect, and give a survey of the most important non-repudiation protocols without and with trusted third party (TTP). For the later ones we discuss the evolution of the TTP's involvement and, between others, describe the most recent protocol using a transparent TTP. We also discuss some ad-hoc problems related to the management of non-repudiation evidences.

Dynamic movement and positioning of embodied agents in multiparty conversations For embodied agents to engage in realistic multiparty conversation, they must stand in appropriate places with respect to other agents and the environment. When these factors change, such as an agent joining the conversation, the agents must dynamically move to a new location and/or orientation to accommodate. This paper presents an algorithm for simulating movement of agents based on observed human behavior using techniques developed for pedestrian movement in crowd simulations. We extend a previous group conversation simulation to include an agent motion algorithm. We examine several test cases and show how the simulation generates results that mirror real-life conversation settings.

Nearly optimal control laws for nonlinear systems with saturating actuators using a neural network HJB approach The Hamilton-Jacobi-Bellman (HJB) equation corresponding to constrained control is formulated using a suitable nonquadratic functional. It is shown that the constrained optimal control law has the largest region of asymptotic stability (RAS). The value function of this HJB equation is solved for by solving for a sequence of cost functions satisfying a sequence of Lyapunov equations (LE). A neural network is used to approximate the cost function associated with each LE using the method of least-squares on a well-defined region of attraction of an initial stabilizing controller. As the order of the neural network is increased, the least-squares solution of the HJB equation converges uniformly to the exact solution of the inherently nonlinear HJB equation associated with the saturating control inputs. The result is a nearly optimal constrained state feedback controller that has been tuned a priori off-line.

Automated Demand Response From Home Energy Management System Under Dynamic Pricing and Power and Comfort Constraints This paper presents a comprehensive and general optimization-based home energy management controller, incorporating several classes of domestic appliances including deferrable, curtailable, thermal, and critical ones. The operations of the appliances are controlled in response to dynamic price signals to reduce the consumer's electricity bill whilst minimizing the daily volume of curtailed energy, and therefore considering the user's comfort level. To avoid shifting a large portion of consumer demand toward the least price intervals, which could create network issues due to loss of diversity, higher prices are applied when the consumer's demand goes beyond a prescribed power threshold. The arising mixed integer nonlinear optimization problem is solved in an iterative manner rolling throughout the day to follow the changes in the anticipated price signals and the variations in the controller inputs while information is updated. The results from different realistic case studies show the effectiveness of the proposed controller in minimizing the household's daily electricity bill while {preserving} comfort level, as well as preventing creation of new least-price peaks.

Scalable and Privacy-Preserving Data Sharing Based on Blockchain. With the development of network technology and cloud computing, data sharing is becoming increasingly popular, and many scholars have conducted in-depth research to promote its flourish. As the scale of data sharing expands, its privacy protection has become a hot issue in research. Moreover, in data sharing, the data is usually maintained in multiple parties, which brings new challenges to protect the privacy of these multi-party data. In this paper, we propose a trusted data sharing scheme using blockchain. We use blockchain to prevent the shared data from being tampered, and use the Paillier cryptosystem to realize the confidentiality of the shared data. In the proposed scheme, the shared data can be traded, and the transaction information is protected by using the (p, t)-threshold Paillier cryptosystem. We conduct experiments in cloud storage scenarios and the experimental results demonstrate the efficiency and effectiveness of the proposed scheme.

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

Collaborative APIs recommendation for Artificial Intelligence of Things with information fusion With the rapid development of Artificial Intelligence of Things (AIoT), many applications are developed and deployed, especially mobile applications and edge applications. Many softwares are developed to support such diverse applications. To facilitate the development of softwares for AIoT, developers and programmers usually rely on the employment of the mature application programming interfaces (APIs). However, it is difficult for developers to select the most suitable APIs to finish the development, decreasing the development efficiency and delaying the development schedule. To solve these problems, in this paper, we propose a collaborative framework for APIs recommendation for AIoT, and also propose a joint matrix factorization technique for information fusion to fully use different types of information in AIoT. The built framework uses the invocation records among users and APIs during the development. Using collaborative technologies, we yield the similarity relationships among users and among APIs and build three novel APIs recommendation models. We collected a real-world dataset and performed sufficient experiments. The experimental results demonstrate that our models produce superior recommendation accuracy.

Delay-Aware Microservice Coordination in Mobile Edge Computing: A Reinforcement Learning Approach As an emerging service architecture, microservice enables decomposition of a monolithic web service into a set of independent lightweight services which can be executed independently. With mobile edge computing, microservices can be further deployed in edge clouds dynamically, launched quickly, and migrated across edge clouds easily, providing better services for users in proximity. However, the user mobility can result in frequent switch of nearby edge clouds, which increases the service delay when users move away from their serving edge clouds. To address this issue, this article investigates microservice coordination among edge clouds to enable seamless and real-time responses to service requests from mobile users. The objective of this work is to devise the optimal microservice coordination scheme which can reduce the overall service delay with low costs. To this end, we first propose a dynamic programming-based offline microservice coordination algorithm, that can achieve the globally optimal performance. However, the offline algorithm heavily relies on the availability of the prior information such as computation request arrivals, time-varying channel conditions and edge cloud's computation capabilities required, which is hard to be obtained. Therefore, we reformulate the microservice coordination problem using Markov decision process framework and then propose a reinforcement learning-based online microservice coordination algorithm to learn the optimal strategy. Theoretical analysis proves that the offline algorithm can find the optimal solution while the online algorithm can achieve near-optimal performance. Furthermore, based on two real-world datasets, i.e., the Telecom's base station dataset and Taxi Track dataset from Shanghai, experiments are conducted. The experimental results demonstrate that the proposed online algorithm outperforms existing algorithms in terms of service delay and migration costs, and the achieved performance is close to the optimal performance obtained by the offline algorithm.

Energy-Aware Task Offloading and Resource Allocation for Time-Sensitive Services in Mobile Edge Computing Systems Mobile Edge Computing (MEC) is a promising architecture to reduce the energy consumption of mobile devices and provide satisfactory quality-of-service to time-sensitive services. How to jointly optimize task offloading and resource allocation to minimize the energy consumption subject to the latency requirement remains an open problem, which motivates this paper. When the latency constraint is tak...

Dynamic Deployment and Cost-Sensitive Provisioning for Elastic Mobile Cloud Services. As mobile customers gradually occupying the largest share of cloud service users, the effective and cost-sensitive provisioning of mobile cloud services quickly becomes a main theme in cloud computing. The key issues involved are much more than just enabling mobile users to access remote cloud resources through wireless networks. The resource limited and intermittent disconnection problems of mobile environments have intrinsic conflict with the continuous connection assumption of the cloud service usage patterns. We advocate that seamless service provisioning in mobile cloud can only be achieved with full exploitation of all available resources around mobile users. An elastic framework is proposed to automatically and dynamically deploy cloud services on data center, base stations, client units, even peer devices. The best deployment location is dynamically determined based on a context-aware and cost-sensitive evaluation model. To facilitate easy adoption of the proposed framework, a service development model and associated semi-automatic tools are provided such that cloud service developers can easily convert a service for execution on different platforms without porting. Prototype implementation and evaluation on the Google Cloud and Android platforms demonstrate that our mechanism can successfully maintain seamless services with very low overhead.

Distributed and Dynamic Service Placement in Pervasive Edge Computing Networks The explosive growth of mobile devices promotes the prosperity of novel mobile applications, which can be realized by service offloading with the assistance of edge computing servers. However, due to limited computation and storage capabilities of a single server, long service latency hinders the continuous development of service offloading in mobile networks. By supporting multi-server cooperation, Pervasive Edge Computing (PEC) is promising to enable service migration in highly dynamic mobile networks. With the objective of maximizing the system utility, we formulate the optimization problem by jointly considering the constraints of server storage capability and service execution latency. To enable dynamic service placement, we first utilize Lyapunov optimization method to decompose the long-term optimization problem into a series of instant optimization problems. Then, a sample average approximation-based stochastic algorithm is proposed to approximate the future expected system utility. Afterwards, a distributed Markov approximation algorithm is utilized to determine the service placement configurations. Through theoretical analysis, the time complexity of our proposed algorithm is linear to the number of users, and the backlog queue of PEC servers is stable. Performance evaluations are conducted based on both synthetic and real trace-driven scenarios, with numerical results demonstrating the effectiveness of our proposed algorithm from various aspects.

A Cooperative Resource Allocation Model For Iot Applications In Mobile Edge Computing With the advancement in the development of the Internet of Things (IoT) technology, as well as the industrial IoT, various applications and services are benefiting from this emerging technology such as smart healthcare systems, virtual realities applications, connected and autonomous vehicles, to name a few. However, IoT devices are known for being limited computation capacities which is crucial to the device's availability time. Traditional approaches used to offload the applications to the cloud to ease the burden on the end user's devices, however, greater latency and network traffic issues still persist. Mobile Edge Computing (MEC) technology has emerged to address these issues and enhance the survivability of cloud infrastructure. While a lot of attempts have been made to manage an efficient process of applications offload, many of which either focus on the allocation of computational or communication protocols without considering a cooperative solution. In addition, a single-user scenario was considered. Therefore, we study multi-user IoT applications offloading for a MEC system, which cooperatively considers to allocate both the resources of computation and communication. The proposed system focuses on minimizing the weighted overhead of local IoT devices, and minimize the offload measured by the delay and energy consumption. The mathematical formulation is a typical mixed integer nonlinear programming (MINP), and this is an NP-hard problem. We obtain the solution to the objective function by splitting the objective problem into three sub-problems. Extensive set of evaluations have been performed so as to get the evaluation of the proposed model. The collected results indicate that offloading decisions, energy consumption, latency, and the impact of the number of IoT devices have shown superior improvement over traditional models.

Probabilistic encryption A new probabilistic model of data encryption is introduced. For this model, under suitable complexity assumptions, it is proved that extracting any information about the cleartext from the cyphertext is hard on the average for an adversary with polynomially bounded computational resources. The proof holds for any message space with any probability distribution. The first implementation of this model is presented. The security of this implementation is proved under the interactability assumptin of deciding Quadratic Residuosity modulo composite numbers whose factorization is unknown.

A powerful and efficient algorithm for numerical function optimization: artificial bee colony (ABC) algorithm Swarm intelligence is a research branch that models the population of interacting agents or swarms that are able to self-organize. An ant colony, a flock of birds or an immune system is a typical example of a swarm system. Bees' swarming around their hive is another example of swarm intelligence. Artificial Bee Colony (ABC) Algorithm is an optimization algorithm based on the intelligent behaviour of honey bee swarm. In this work, ABC algorithm is used for optimizing multivariable functions and the results produced by ABC, Genetic Algorithm (GA), Particle Swarm Algorithm (PSO) and Particle Swarm Inspired Evolutionary Algorithm (PS-EA) have been compared. The results showed that ABC outperforms the other algorithms.

Markov games as a framework for multi-agent reinforcement learning In the Markov decision process (MDP) formalization of reinforcement learning, a single adaptive agent interacts with an environment defined by a probabilistic transition function. In this solipsis-tic view, secondary agents can only be part of the environment and are therefore fixed in their behavior. The framework of Markov games allows us to widen this view to include multiple adaptive agents with interacting or competing goals. This paper considers a step in this direction in which exactly two agents with diametrically opposed goals share an environment. It describes a Q-learning-like algorithm for finding optimal policies and demonstrates its application to a simple two-player game in which the optimal policy is probabilistic.

Scalable and efficient provable data possession. Storage outsourcing is a rising trend which prompts a number of interesting security issues, many of which have been extensively investigated in the past. However, Provable Data Possession (PDP) is a topic that has only recently appeared in the research literature. The main issue is how to frequently, efficiently and securely verify that a storage server is faithfully storing its client's (potentially very large) outsourced data. The storage server is assumed to be untrusted in terms of both security and reliability. (In other words, it might maliciously or accidentally erase hosted data; it might also relegate it to slow or off-line storage.) The problem is exacerbated by the client being a small computing device with limited resources. Prior work has addressed this problem using either public key cryptography or requiring the client to outsource its data in encrypted form. In this paper, we construct a highly efficient and provably secure PDP technique based entirely on symmetric key cryptography, while not requiring any bulk encryption. Also, in contrast with its predecessors, our PDP technique allows outsourcing of dynamic data, i.e, it efficiently supports operations, such as block modification, deletion and append.

Cognitive Cars: A New Frontier for ADAS Research This paper provides a survey of recent works on cognitive cars with a focus on driver-oriented intelligent vehicle motion control. The main objective here is to clarify the goals and guidelines for future development in the area of advanced driver-assistance systems (ADASs). Two major research directions are investigated and discussed in detail: 1) stimuli–decisions–actions, which focuses on the driver side, and 2) perception enhancement–action-suggestion–function-delegation, which emphasizes the ADAS side. This paper addresses the important achievements and major difficulties of each direction and discusses how to combine the two directions into a single integrated system to obtain safety and comfort while driving. Other related topics, including driver training and infrastructure design, are also studied.

Online Prediction of Driver Distraction Based on Brain Activity Patterns This paper presents a new computational framework for early detection of driver distractions (map viewing) using brain activity measured by electroencephalographic (EEG) signals. Compared with most studies in the literature, which are mainly focused on the classification of distracted and nondistracted periods, this study proposes a new framework to prospectively predict the start and end of a distraction period, defined by map viewing. The proposed prediction algorithm was tested on a data set of continuous EEG signals recorded from 24 subjects. During the EEG recordings, the subjects were asked to drive from an initial position to a destination using a city map in a simulated driving environment. The overall accuracy values for the prediction of the start and the end of map viewing were 81% and 70%, respectively. The experimental results demonstrated that the proposed algorithm can predict the start and end of map viewing with relatively high accuracy and can be generalized to individual subjects. The outcome of this study has a high potential to improve the design of future intelligent navigation systems. Prediction of the start of map viewing can be used to provide route information based on a driver's needs and consequently avoid map-viewing activities. Prediction of the end of map viewing can be used to provide warnings for potential long map-viewing durations. Further development of the proposed framework and its applications in driver-distraction predictions are also discussed.

Adaptive fuzzy tracking control for switched uncertain strict-feedback nonlinear systems. •Adaptive tracking control for switched strict-feedback nonlinear systems is proposed.•The generalized fuzzy hyperbolic model is used to approximate nonlinear functions.•The designed controller has fewer design parameters comparing with existing methods.

Social Robots for (Second) Language Learning in (Migrant) Primary School Children Especially these days, innovation and support from technology to relieve pressure in education is highly urgent. This study tested the potential advantage of a social robot over a tablet in (second) language learning on performance, engagement, and enjoyment. Shortages in primary education call for new technology solutions. Previous studies combined robots with tablets, to compensate for robot’s limitations, however, this study applied direct human–robot interaction. Primary school children (N = 63, aged 4–6) participated in a 3-wave field experiment with story-telling exercises, either with a semi-autonomous robot (without tablet, using WOz) or a tablet. Results showed increased learning gains over time when training with a social robot, compared to the tablet. Children who trained with a robot were more engaged in the story-telling task and enjoyed it more. Robot’s behavioral style (social or neutral) hardly differed overall, however, seems to vary for high versus low educational abilities. While social robots need sophistication before being implemented in schools, our study shows the potential of social robots as tutors in (second) language learning.

Vehicle Detection and Disparity Estimation Using Blended Stereo Images In this article, a new method is presented for simultaneous vehicle detection and disparity estimation from the blended multiple-view image. The multiple-view image is created by blending left and right images from a stereo camera. This blended image can implicitly encode the appearance and view disparity of an object in a single image, which provides clues for detection and disparity estimation. ...

Generative Adversarial Networks for Parallel Transportation Systems. Generative Adversaria Networks (GANs) have emerged as a promising and effective mechanism for machine learning due to its recent successful applications. GANs share the same idea of producing, testing, acquiring, and utilizing data as well as knowledge based on artificial systems, computational experiments, and parallel execution of actual and virtual scenarios, as outlined in the theory of parall...

Deep Multi-Modal Object Detection and Semantic Segmentation for Autonomous Driving: Datasets, Methods, and Challenges AbstractRecent advancements in perception for autonomous driving are driven by deep learning. In order to achieve robust and accurate scene understanding, autonomous vehicles are usually equipped with different sensors (e.g. cameras, LiDARs, Radars), and multiple sensing modalities can be fused to exploit their complementary properties. In this context, many methods have been proposed for deep multi-modal perception problems. However, there is no general guideline for network architecture design, and questions of “what to fuse”, “when to fuse”, and “how to fuse” remain open. This review paper attempts to systematically summarize methodologies and discuss challenges for deep multi-modal object detection and semantic segmentation in autonomous driving. To this end, we first provide an overview of on-board sensors on test vehicles, open datasets, and background information for object detection and semantic segmentation in autonomous driving research. We then summarize the fusion methodologies and discuss challenges and open questions. In the appendix, we provide tables that summarize topics and methods. We also provide an interactive online platform to navigate each reference: https://boschresearch.github.io/multimodalperception/.

Enhanced Object Detection With Deep Convolutional Neural Networks for Advanced Driving Assistance Object detection is a critical problem for advanced driving assistance systems (ADAS). Recently, convolutional neural networks (CNN) achieved large successes on object detection, with performance improvement over traditional approaches, which use hand-engineered features. However, due to the challenging driving environment (e.g., large object scale variation, object occlusion, and bad light conditions), popular CNN detectors do not achieve very good object detection accuracy over the KITTI autonomous driving benchmark dataset. In this paper, we propose three enhancements for CNN-based visual object detection for ADAS. To address the large object scale variation challenge, deconvolution and fusion of CNN feature maps are proposed to add context and deeper features for better object detection at low feature map scales. In addition, soft non-maximal suppression (NMS) is applied across object proposals at different feature scales to address the object occlusion challenge. As the cars and pedestrians have distinct aspect ratio features, we measure their aspect ratio statistics and exploit them to set anchor boxes properly for better object matching and localization. The proposed CNN enhancements are evaluated with various image input sizes by experiments over KITTI dataset. The experimental results demonstrate the effectiveness of the proposed enhancements with good detection performance over KITTI test set.

MFR-CNN: Incorporating Multi-Scale Features and Global Information for Traffic Object Detection. Object detection plays an important role in intelligent transportation systems and intelligent vehicles. Although the topic of object detection has been studied for decades, it is still challenging to accurately detect objects under complex scenarios. The contributing factors for challenges include diversified object and background appearance, motion blur, adverse weather conditions, and complex i...

Context-Aware Dynamic Feature Extraction for 3D Object Detection in Point Clouds Varying density of point clouds increases the difficulty of 3D detection. In this paper, we present a context-aware dynamic network (CADNet) to capture the variance of density by considering both point context and semantic context. Point-level contexts are generated from original point clouds to enlarge the effective receptive filed. They are extracted around the voxelized pillars based on our extended voxelization method and processed with the context encoder in parallel with the pillar features. With a large perception range, we are able to capture the variance of features for potential objects and generate attentive spatial guidance to help adjust the strengths for different regions. In the region proposal network, considering the limited representation ability of traditional convolution where same kernels are shared among different samples and positions, we propose a decomposable dynamic convolutional layer to adapt to the variance of input features by learning from the local semantic context. It adaptively generates the position-dependent coefficients for multiple fixed kernels and combines them to convolve with local features. Based on our dynamic convolution, we design a dual-path convolution block to further improve the representation ability. We conduct experiments on KITTI dataset and the proposed CADNet has achieved superior performance of 3D detection outperforming SECOND and PointPillars by a large margin at the speed of 30 FPS.

China's 12-Year Quest of Autonomous Vehicular Intelligence: The Intelligent Vehicles Future Challenge Program In this article, we introduce the Intelligent Vehicles Future Challenge of China (IVFC), which has lasted 12 years. Some key features of the tests and a few interesting findings of IVFC are selected and presented. Through the IVFCs held between 2009 and 2020, we gradually established a set of theories, methods, and tools to collect tests? data and efficiently evaluate the performance of autonomous vehicles so that we could learn how to improve both the autonomous vehicles and the testing system itself.

Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation Object detection performance, as measured on the canonical PASCAL VOC dataset, has plateaued in the last few years. The best-performing methods are complex ensemble systems that typically combine multiple low-level image features with high-level context. In this paper, we propose a simple and scalable detection algorithm that improves mean average precision (mAP) by more than 30% relative to the previous best result on VOC 2012 -- achieving a mAP of 53.3%. Our approach combines two key insights: (1) one can apply high-capacity convolutional neural networks (CNNs) to bottom-up region proposals in order to localize and segment objects and (2) when labeled training data is scarce, supervised pre-training for an auxiliary task, followed by domain-specific fine-tuning, yields a significant performance boost. Since we combine region proposals with CNNs, we call our method R-CNN: Regions with CNN features. We also present experiments that provide insight into what the network learns, revealing a rich hierarchy of image features. Source code for the complete system is available at http://www.cs.berkeley.edu/~rbg/rcnn.

A comparative study of texture measures with classification based on featured distributions This paper evaluates the performance both of some texture measures which have been successfully used in various applications and of some new promising approaches proposed recently. For classification a method based on Kullback discrimination of sample and prototype distributions is used. The classification results for single features with one-dimensional feature value distributions and for pairs of complementary features with two-dimensional distributions are presented

Social Perception and Steering for Online Avatars This paper presents work on a new platform for producing realistic group conversation dynamics in shared virtual environments. An avatar, representing users, should perceive the surrounding social environment just as humans would, and use the perceptual information for driving low level reactive behaviors. Unconscious reactions serve as evidence of life, and can also signal social availability and spatial awareness to others. These behaviors get lost when avatar locomotion requires explicit user control. For automating such behaviors we propose a steering layer in the avatars that manages a set of prioritized behaviors executed at different frequencies, which can be activated or deactivated and combined together. This approach gives us enough flexibility to model the group dynamics of social interactions as a set of social norms that activate relevant steering behaviors. A basic set of behaviors is described for conversations, some of which generate a social force field that makes the formation of conversation groups fluidly adapt to external and internal noise, through avatar repositioning and reorientations. The resulting social group behavior appears relatively robust, but perhaps more importantly, it starts to bring a new sense of relevance and continuity to the virtual bodies that often get separated from the ongoing conversation in the chat window.

Node Reclamation and Replacement for Long-Lived Sensor Networks When deployed for long-term tasks, the energy required to support sensor nodes' activities is far more than the energy that can be preloaded in their batteries. No matter how the battery energy is conserved, once the energy is used up, the network life terminates. Therefore, guaranteeing long-term energy supply has persisted as a big challenge. To address this problem, we propose a node reclamation and replacement (NRR) strategy, with which a mobile robot or human labor called mobile repairman (MR) periodically traverses the sensor network, reclaims nodes with low or no power supply, replaces them with fully charged ones, and brings the reclaimed nodes back to an energy station for recharging. To effectively and efficiently realize the strategy, we develop an adaptive rendezvous-based two-tier scheduling scheme (ARTS) to schedule the replacement/reclamation activities of the MR and the duty cycles of nodes. Extensive simulations have been conducted to verify the effectiveness and efficiency of the ARTS scheme.

Haptic feedback for enhancing realism of walking simulations. In this paper, we describe several experiments whose goal is to evaluate the role of plantar vibrotactile feedback in enhancing the realism of walking experiences in multimodal virtual environments. To achieve this goal we built an interactive and a noninteractive multimodal feedback system. While during the use of the interactive system subjects physically walked, during the use of the noninteractive system the locomotion was simulated while subjects were sitting on a chair. In both the configurations subjects were exposed to auditory and audio-visual stimuli presented with and without the haptic feedback. Results of the experiments provide a clear preference toward the simulations enhanced with haptic feedback showing that the haptic channel can lead to more realistic experiences in both interactive and noninteractive configurations. The majority of subjects clearly appreciated the added feedback. However, some subjects found the added feedback unpleasant. This might be due, on one hand, to the limits of the haptic simulation and, on the other hand, to the different individual desire to be involved in the simulations. Our findings can be applied to the context of physical navigation in multimodal virtual environments as well as to enhance the user experience of watching a movie or playing a video game.

Vehicular Sensing Networks in a Smart City: Principles, Technologies and Applications. Given the escalating population across the globe, it has become paramount to construct smart cities, aiming for improving the management of urban flows relying on efficient information and communication technologies (ICT). Vehicular sensing networks (VSNs) play a critical role in maintaining the efficient operation of smart cities. Naturally, there are numerous challenges to be solved before the w...

Dual-objective mixed integer linear program and memetic algorithm for an industrial group scheduling problem Group scheduling problems have attracted much attention owing to their many practical applications. This work proposes a new bi-objective serial-batch group scheduling problem considering the constraints of sequence-dependent setup time, release time, and due time. It is originated from an important industrial process, i.e., wire rod and bar rolling process in steel production systems. Two objecti...

One billion word benchmark for measuring progress in statistical language modeling. We propose a new benchmark corpus to be used for measuring progress in statistical language modeling. With almost one billion words of training data, we hope this benchmark will be useful to quickly evaluate novel language modeling techniques, and to compare their contribution when combined with other advanced techniques. We show performance of several well-known types of language models, with the best results achieved with a recurrent neural network based language model. The baseline unpruned Kneser-Ney 5-gram model achieves perplexity 67.6; a combination of techniques leads to 35% reduction in perplexity, or 10% reduction in cross-entropy (bits), over that baseline. The benchmark is available as a code.google.com project; besides the scripts needed to rebuild the training/held-out data, it also makes available log-probability values for each word in each of ten held-out data sets, for each of the baseline n-gram models.

Adversarial Example Generation with Syntactically Controlled Paraphrase Networks. We propose syntactically controlled paraphrase networks (SCPNs) and use them to generate adversarial examples. Given a sentence and a target syntactic form (e.g., a constituency parse), SCPNs are trained to produce a paraphrase of the sentence with the desired syntax. We show it is possible to create training data for this task by first doing backtranslation at a very large scale, and then using a parser to label the syntactic transformations that naturally occur during this process. Such data allows us to train a neural encoder-decoder model with extra inputs to specify the target syntax. A combination of automated and human evaluations show that SCPNs generate paraphrases that follow their target specifications without decreasing paraphrase quality when compared to baseline (uncontrolled) paraphrase systems. Furthermore, they are more capable of generating syntactically adversarial examples that both (1) fool pretrained models and (2) improve the robustness of these models to syntactic variation when used to augment their training data.

GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding. For natural language understanding (NLU) technology to be maximally useful, both practically and as a scientific object of study, it must be general: it must be able to process language in a way that is not exclusively tailored to any one specific task or dataset. In pursuit of this objective, we introduce the General Language Understanding Evaluation benchmark (GLUE), a tool for evaluating and analyzing the performance of models across a diverse range of existing NLU tasks. GLUE is model-agnostic, but it incentivizes sharing knowledge across tasks because certain tasks have very limited training data. We further provide a hand-crafted diagnostic test suite that enables detailed linguistic analysis of NLU models. We evaluate baselines based on current methods for multi-task and transfer learning and find that they do not immediately give substantial improvements over the aggregate performance of training a separate model per task, indicating room for improvement in developing general and robust NLU systems.

Hitting the right paraphrases in good time We present a random-walk-based approach to learning paraphrases from bilingual parallel corpora. The corpora are represented as a graph in which a node corresponds to a phrase, and an edge exists between two nodes if their corresponding phrases are aligned in a phrase table. We sample random walks to compute the average number of steps it takes to reach a ranking of paraphrases with better ones being "closer" to a phrase of interest. This approach allows "feature" nodes that represent domain knowledge to be built into the graph, and incorporates truncation techniques to prevent the graph from growing too large for efficiency. Current approaches, by contrast, implicitly presuppose the graph to be bipartite, are limited to finding paraphrases that are of length two away from a phrase, and do not generally permit easy incorporation of domain knowledge. Manual evaluation of generated output shows that our approach outperforms the state-of-the-art system of Callison-Burch (2008).

Style Transfer in Text: Exploration and Evaluation The ability to transfer styles of texts or images, is an important measurement of the advancement of artificial intelligence (AI). However, the progress in language style transfer is lagged behind other domains, such as computer vision, mainly because of the lack of parallel data and reliable evaluation metrics. in response to the challenge of lacking parallel data, we explore learning style transfer from non-parallel data. We propose two models to achieve this goal. The key idea behind the proposed models is to learn separate content representations and style representations using adversarial networks. Considering the problem of lacking principle evaluation metrics, we propose two novel evaluation metrics that measure two aspects of style transfer: transfer strength and content preservation. We benchmark our models and the evaluation metrics on two style transfer tasks: paper-news title transfer, and positive-negative review transfer. Results show that the proposed content preservation metric is highly correlate to human judgments, and the proposed models are able to generate sentences with similar content preservation score but higher style transfer strength comparing to auto encoder.

A Multi-Task Approach for Disentangling Syntax and Semantics in Sentence Representations. We propose a generative model for a sentence that uses two latent variables, with one intended to represent the syntax of the sentence and the other to represent its semantics. We show we can achieve better disentanglement between semantic and syntactic representations by training with multiple losses, including losses that exploit aligned paraphrastic sentences and word-order information. We also investigate the effect of moving from bag-of-words to recurrent neural network modules. We evaluate our models as well as several popular pretrained embeddings on standard semantic similarity tasks and novel syntactic similarity tasks. Empirically, we find that the model with the best performing syntactic and semantic representations also gives rise to the most disentangled representations.

Glove: Global Vectors for Word Representation.

Adam: A Method for Stochastic Optimization. We introduce Adam, an algorithm for first-order gradient-based optimization of stochastic objective functions, based on adaptive estimates of lower-order moments. The method is straightforward to implement, is computationally efficient, has little memory requirements, is invariant to diagonal rescaling of the gradients, and is well suited for problems that are large in terms of data and/or parameters. The method is also appropriate for non-stationary objectives and problems with very noisy and/or sparse gradients. The hyper-parameters have intuitive interpretations and typically require little tuning. Some connections to related algorithms, on which Adam was inspired, are discussed. We also analyze the theoretical convergence properties of the algorithm and provide a regret bound on the convergence rate that is comparable to the best known results under the online convex optimization framework. Empirical results demonstrate that Adam works well in practice and compares favorably to other stochastic optimization methods. Finally, we discuss AdaMax, a variant of Adam based on the infinity norm.

Rethinking The Inception Architecture For Computer Vision Convolutional networks are at the core of most state-of-the-art computer vision solutions for a wide variety of tasks. Since 2014 very deep convolutional networks started to become mainstream, yielding substantial gains in various benchmarks. Although increased model size and computational cost tend to translate to immediate quality gains for most tasks (as long as enough labeled data is provided for training), computational efficiency and low parameter count are still enabling factors for various use cases such as mobile vision and big-data scenarios. Here we are exploring ways to scale up networks in ways that aim at utilizing the added computation as efficiently as possible by suitably factorized convolutions and aggressive regularization. We benchmark our methods on the ILSVRC 2012 classification challenge validation set demonstrate substantial gains over the state of the art: 21.2% top-1 and 5.6% top-5 error for single frame evaluation using a network with a computational cost of 5 billion multiply-adds per inference and with using less than 25 million parameters. With an ensemble of 4 models and multi-crop evaluation, we report 3.5% top-5 error and 17.3% top-1 error on the validation set and 3.6% top-5 error on the official test set.

Changes in the Correlation Between Eye and Steering Movements Indicate Driver Distraction Driver distraction represents an increasingly important contributor to crashes and fatalities. Technology that can detect and mitigate distraction by alerting distracted drivers could play a central role in maintaining safety. Based on either eye measures or driver performance measures, numerous algorithms to detect distraction have been developed. Combining both eye glance and vehicle data could enhance distraction detection. The goal of this paper is to evaluate whether changes in the eye–steering correlation structure can indicate distraction. Drivers performed visual, cognitive, and cognitive/visual tasks while driving in a simulator. The auto- and cross-correlations of horizontal eye position and steering wheel angle show that eye movements associated with road scanning produce a low eye–steering correlation. However, even this weak correlation is sensitive to distraction. Time lead associated with the maximum correlation is sensitive to all three types of distraction, and the maximum correlation coefficient is most strongly affected by off-road glances. These results demonstrate that eye–steering correlation statistics can detect distraction and differentiate between types of distraction.

Stabilizing a linear system by switching control with dwell time The use of networks in control systems to connect controllers and sensors/actuators has become common practice in many applications. This new technology has also posed a theoretical control problem of how to use the limited data rate of the network effectively. We consider a system where its sensor and actuator are connected by a finite data rate channel. A design method to stabilize a continuous-time, linear plant using a switching controller is proposed. In particular, to prevent the actuator from fast switching, or chattering, which can not only increase the necessary data rate but also damage the system, we employ a dwell-time switching scheme. It is shown that a systematic partition of the state-space enables us to reduce the complexity of the design problem

A Survey on Vehicle Routing Problem with Loading Constraints In the classical vehicle routing problem (VRP), a fleet of vehicles is available to serve a set of customers with known demand. Each customer is visited by exactly one vehicle as well as the objective is to minimize the total distance or the total charge incurred. Recent years have seen increased attention on VRP integrated with additional loading constraints, known as 2L-CVRP or 3L-CVRP. In this paper we present a survey of the state-of-the-art on 2L-CVRP/3L-CVRP.

Real-Time Video Analytics: The Killer App for Edge Computing. Video analytics will drive a wide range of applications with great potential to impact society. A geographically distributed architecture of public clouds and edges that extend down to the cameras is the only feasible approach to meeting the strict real-time requirements of large-scale live video analytics.

Social Robots for (Second) Language Learning in (Migrant) Primary School Children Especially these days, innovation and support from technology to relieve pressure in education is highly urgent. This study tested the potential advantage of a social robot over a tablet in (second) language learning on performance, engagement, and enjoyment. Shortages in primary education call for new technology solutions. Previous studies combined robots with tablets, to compensate for robot’s limitations, however, this study applied direct human–robot interaction. Primary school children (N = 63, aged 4–6) participated in a 3-wave field experiment with story-telling exercises, either with a semi-autonomous robot (without tablet, using WOz) or a tablet. Results showed increased learning gains over time when training with a social robot, compared to the tablet. Children who trained with a robot were more engaged in the story-telling task and enjoyed it more. Robot’s behavioral style (social or neutral) hardly differed overall, however, seems to vary for high versus low educational abilities. While social robots need sophistication before being implemented in schools, our study shows the potential of social robots as tutors in (second) language learning.

Dynamic transfer among alternative controllers and its relation to antiwindup controller design Advanced control strategies and modern consulting provide new challenges for the classical problem of bumpless transfer. It can, for example, be necessary to transfer between an only approximately known existing analog controller and a new digital or adaptive controller without accessing any states. Transfer ought to be bidirectional and not presuppose steady state, so that an immediate back-transfer is possible if the new controller should drive the plant unstable. We present a scheme that meets these requirements. By casting the problem of bidirectional transfer into an associated tracking control problem, systematic analysis and design procedures from control theory can be applied. The associated control problem also has a correspondence to the design of antiwindup controllers. The paper includes laboratory and industrial applications.

State resetting for bumpless switching in supervisory control In this paper the realization and implementation of a multi-controller scheme made of a finite set of linear single-input-single-output controllers, possibly having different state dimensions, is studied. The supervisory control framework is considered, namely a minimal parameter dependent realization of the set of controllers such that all controllers share the same state space is used. A specific state resetting strategy based on the behavioral approach to system theory is developed in order to master the transient upon controller switching.

Interpolation for gain-scheduled control with guarantees Here, a methodology is presented which considers the interpolation of linear time-invariant (LTI) controllers designed for different operating points of a nonlinear system in order to produce a gain-scheduled controller. Guarantees of closed-loop quadratic stability and performance at intermediate interpolation points are presented in terms of a set of linear matrix inequalities (LMIs). The proposed interpolation scheme can be applied in cases where the system must remain at the operating points most of the time and the transitions from one point to another rarely occur, e.g., chemical processes, satellites.

On controller initialization in multivariable switching systems We consider a class of switched systems which consists of a linear MIMO and possibly unstable process in feedback interconnection with a multicontroller whose dynamics switch. It is shown how one can achieve significantly better transient performance by selecting the initial condition for every controller when it is inserted into the feedback loop. This initialization is obtained by performing the minimization of a quadratic cost function of the tracking error, controlled output, and control signal. We guarantee input-to-state stability of the closed-loop system when the average number of switches per unit of time is smaller than a specific value. If this is not the case then stability can still be achieved by adding a mild constraint to the optimization. We illustrate the use of our results in the control of a flexible beam actuated in torque. This system is unstable with two poles at the origin and contains several lightly damped modes, which can be easily excited by controller switching.

A simple approach for switched control design with control bumps limitation. By its own nature, control of switched systems in general leads to expressive discontinuities at switching times. Hence, this class of dynamic systems needs additional care as far as implementation constraints such as for instance control amplitude limitation is concerned. This paper aims at providing numerically tractable conditions to be incorporated in the control design procedure in order to reduce control bumps. The switching strategy and continuous control laws are jointly determined as well as an H∞ guaranteed cost is minimized. Due to its theoretical and practical importance, special attention is given to the dynamic output feedback control design problem. The results are illustrated by means of examples borrowed from the literature which are also used for comparisons that put in evidence the efficiency of the proposed strategy.

A parameter set division and switching gain-scheduling controllers design method for time-varying plants. This paper presents a new technique to design switching gain-scheduling controllers for plants with measurable time-varying parameters. By dividing the parameter set into a sufficient number of subsets, and by designing a robust controller to each subset, the designed switching gain-scheduling controllers achieve a desired L2-gain performance for each subset, while ensuring stability whenever a controller switching occurs due to the crossing of the time-varying parameters between any two adjacent subsets. Based on integral quadratic constraints theory and Lyapunov stability theory, a switching gain-scheduling controllers design problem amounts to solving optimization problems. Each optimization problem is to be solved by a combination of the bisection search and the numerical nonsmooth optimization method. The main advantage of the proposed technique is that the division of the parameter region is determined automatically, without any prespecified parameter set division which is required in most of previously developed switching gain-scheduling controllers design methods. A numerical example illustrates the validity of the proposed technique.

Convexity of the cost functional in an optimal control problem for a class of positive switched systems. This paper deals with the optimal control of a class of positive switched systems. The main feature of this class is that switching alters only the diagonal entries of the dynamic matrix. The control input is represented by the switching signal itself and the optimal control problem is that of minimizing a positive linear combination of the final state variable. First, the switched system is embedded in the class of bilinear systems with control variables living in a simplex, for each time point. The main result is that the cost is convex with respect to the control variables. This ensures that any Pontryagin solution is optimal. Algorithms to find the optimal solution are then presented and an example, taken from a simplified model for HIV mutation mitigation is discussed.

Hamming Embedding and Weak Geometric Consistency for Large Scale Image Search This paper improves recent methods for large scale image search. State-of-the-art methods build on the bag-of-features image representation. We, first, analyze bag-of-features in the framework of approximate nearest neighbor search. This shows the sub-optimality of such a representation for matching descriptors and leads us to derive a more precise representation based on 1) Hamming embedding (HE) and 2) weak geometric consistency constraints (WGC). HE provides binary signatures that refine the matching based on visual words. WGC filters matching descriptors that are not consistent in terms of angle and scale. HE and WGC are integrated within the inverted file and are efficiently exploited for all images, even in the case of very large datasets. Experiments performed on a dataset of one million of images show a significant improvement due to the binary signature and the weak geometric consistency constraints, as well as their efficiency. Estimation of the full geometric transformation, i.e., a re-ranking step on a short list of images, is complementary to our weak geometric consistency constraints and allows to further improve the accuracy.

Delay-Aware Microservice Coordination in Mobile Edge Computing: A Reinforcement Learning Approach As an emerging service architecture, microservice enables decomposition of a monolithic web service into a set of independent lightweight services which can be executed independently. With mobile edge computing, microservices can be further deployed in edge clouds dynamically, launched quickly, and migrated across edge clouds easily, providing better services for users in proximity. However, the user mobility can result in frequent switch of nearby edge clouds, which increases the service delay when users move away from their serving edge clouds. To address this issue, this article investigates microservice coordination among edge clouds to enable seamless and real-time responses to service requests from mobile users. The objective of this work is to devise the optimal microservice coordination scheme which can reduce the overall service delay with low costs. To this end, we first propose a dynamic programming-based offline microservice coordination algorithm, that can achieve the globally optimal performance. However, the offline algorithm heavily relies on the availability of the prior information such as computation request arrivals, time-varying channel conditions and edge cloud's computation capabilities required, which is hard to be obtained. Therefore, we reformulate the microservice coordination problem using Markov decision process framework and then propose a reinforcement learning-based online microservice coordination algorithm to learn the optimal strategy. Theoretical analysis proves that the offline algorithm can find the optimal solution while the online algorithm can achieve near-optimal performance. Furthermore, based on two real-world datasets, i.e., the Telecom's base station dataset and Taxi Track dataset from Shanghai, experiments are conducted. The experimental results demonstrate that the proposed online algorithm outperforms existing algorithms in terms of service delay and migration costs, and the achieved performance is close to the optimal performance obtained by the offline algorithm.

Grey Wolf Optimizer. This work proposes a new meta-heuristic called Grey Wolf Optimizer (GWO) inspired by grey wolves (Canis lupus). The GWO algorithm mimics the leadership hierarchy and hunting mechanism of grey wolves in nature. Four types of grey wolves such as alpha, beta, delta, and omega are employed for simulating the leadership hierarchy. In addition, the three main steps of hunting, searching for prey, encircling prey, and attacking prey, are implemented. The algorithm is then benchmarked on 29 well-known test functions, and the results are verified by a comparative study with Particle Swarm Optimization (PSO), Gravitational Search Algorithm (GSA), Differential Evolution (DE), Evolutionary Programming (EP), and Evolution Strategy (ES). The results show that the GWO algorithm is able to provide very competitive results compared to these well-known meta-heuristics. The paper also considers solving three classical engineering design problems (tension/compression spring, welded beam, and pressure vessel designs) and presents a real application of the proposed method in the field of optical engineering. The results of the classical engineering design problems and real application prove that the proposed algorithm is applicable to challenging problems with unknown search spaces.

Comment on "On Discriminative vs. Generative Classifiers: A Comparison of Logistic Regression and Naive Bayes" Comparison of generative and discriminative classifiers is an ever-lasting topic. As an important contribution to this topic, based on their theoretical and empirical comparisons between the naïve Bayes classifier and linear logistic regression, Ng and Jordan (NIPS 841---848, 2001) claimed that there exist two distinct regimes of performance between the generative and discriminative classifiers with regard to the training-set size. In this paper, our empirical and simulation studies, as a complement of their work, however, suggest that the existence of the two distinct regimes may not be so reliable. In addition, for real world datasets, so far there is no theoretically correct, general criterion for choosing between the discriminative and the generative approaches to classification of an observation x into a class y; the choice depends on the relative confidence we have in the correctness of the specification of either p(y|x) or p(x, y) for the data. This can be to some extent a demonstration of why Efron (J Am Stat Assoc 70(352):892---898, 1975) and O'Neill (J Am Stat Assoc 75(369):154---160, 1980) prefer normal-based linear discriminant analysis (LDA) when no model mis-specification occurs but other empirical studies may prefer linear logistic regression instead. Furthermore, we suggest that pairing of either LDA assuming a common diagonal covariance matrix (LDA-驴) or the naïve Bayes classifier and linear logistic regression may not be perfect, and hence it may not be reliable for any claim that was derived from the comparison between LDA-驴 or the naïve Bayes classifier and linear logistic regression to be generalised to all generative and discriminative classifiers.

GROPING: Geomagnetism and cROwdsensing Powered Indoor NaviGation Although a large number of WiFi fingerprinting based indoor localization systems have been proposed, our field experience with Google Maps Indoor (GMI), the only system available for public testing, shows that it is far from mature for indoor navigation. In this paper, we first report our field studies with GMI, as well as experiment results aiming to explain our unsatisfactory GMI experience. Then motivated by the obtained insights, we propose GROPING as a self-contained indoor navigation system independent of any infrastructural support. GROPING relies on geomagnetic fingerprints that are far more stable than WiFi fingerprints, and it exploits crowdsensing to construct floor maps rather than expecting individual venues to supply digitized maps. Based on our experiments with 20 participants in various floors of a big shopping mall, GROPING is able to deliver a sufficient accuracy for localization and thus provides smooth navigation experience.

Scalable and Privacy-Preserving Data Sharing Based on Blockchain. With the development of network technology and cloud computing, data sharing is becoming increasingly popular, and many scholars have conducted in-depth research to promote its flourish. As the scale of data sharing expands, its privacy protection has become a hot issue in research. Moreover, in data sharing, the data is usually maintained in multiple parties, which brings new challenges to protect the privacy of these multi-party data. In this paper, we propose a trusted data sharing scheme using blockchain. We use blockchain to prevent the shared data from being tampered, and use the Paillier cryptosystem to realize the confidentiality of the shared data. In the proposed scheme, the shared data can be traded, and the transaction information is protected by using the (p, t)-threshold Paillier cryptosystem. We conduct experiments in cloud storage scenarios and the experimental results demonstrate the efficiency and effectiveness of the proposed scheme.

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

NetVLAD: CNN architecture for weakly supervised place recognition We tackle the problem of large scale visual place recognition, where the task is to quickly and accurately recognize the location of a given query photograph. We present the following four principal contributions. First, we develop a convolutional neural network (CNN) architecture that is trainable in an end-to-end manner directly for the place recognition task. The main component of this architec...

CellSense: A Probabilistic RSSI-Based GSM Positioning System Context-aware applications have been gaining huge interest in the last few years. With cell phones becoming ubiquitous computing devices, cell phone localization has become an important research problem. In this paper, we present CellSense, a probabilistic RSSI-based fingerprinting location determination system for GSM phones. We discuss the challenges of implementing a probabilistic fingerprinting localization technique in GSM networks and present the details of the CellSense system and how it addresses the challenges. To evaluate our proposed system, we implemented CellSense on Android-based phones. Results for two different testbeds, representing urban and rural environments, show that CellSense provides at least 23.8% enhancement in accuracy in rural areas and at least 86.4% in urban areas compared to other RSSI-based GSM localization systems. This comes with a minimal increase in computational requirements. We also evaluate the effect of changing the different system parameters on the accuracy-complexity tradeoff.

Using mobile phone barometer for low-power transportation context detection Accelerometer is the predominant sensor used for low-power context detection on smartphones. Although low-power, accelerometer is orientation and position-dependent, requires a high sampling rate, and subsequently complex processing and training to achieve good accuracy. We present an alternative approach for context detection using only the smartphone's barometer, a relatively new sensor now present in an increasing number of devices. The barometer is independent of phone position and orientation. Using a low sampling rate of 1 Hz, and simple processing based on intuitive logic, we demonstrate that it is possible to use the barometer for detecting the basic user activities of IDLE, WALKING, and VEHICLE at extremely low-power. We evaluate our approach using 47 hours of real-world transportation traces from 3 countries and 13 individuals, as well as more than 900 km of elevation data pulled from Google Maps from 5 cities, comparing power and accuracy to Google's accelerometer-based Activity Recognition algorithm, and to Future Urban Mobility Survey's (FMS) GPS-accelerometer server-based application. Our barometer-based approach uses 32 mW lower power compared to Google, and has comparable accuracy to both Google and FMS. This is the first paper that uses only the barometer for context detection.

TDOA-based passive localization of standard WiFi devices Indoor location-based service has widespread applications. With the ubiquitous deployment of WiFi systems, it is of significant interest to provide location-based service using standard WiFi devices. Most of the existing WiFi-based localization techniques are based on Received Signal Strength (RSS) measurements. As the bandwidth of WiFi systems increases, it is possible to achieve accurate timing-based positioning. This work presents a WiFi-based positioning system that has been developed at CSIRO as a research platform, where target devices are located using passive sniffers that measure the Time Difference of Arrival (TDOA) of the packets transmitted by the target devices. This work describes the architecture, hardware, and algorithms of the system, including the techniques used for clock synchronizing and system calibration. It is shown experimentally that the positioning error is 23 cm in open spaces and 1.5 m in an indoor office environment for a 80MHz WiFi system. The system can be used to track standard WiFi devices passively without interfering with the existing WiFi infrastructure, and is ideal for security applications.

Scalable 6-Dof Localization On Mobile Devices Recent improvements in image-based localization have produced powerful methods that scale up to the massive 3D models emerging from modern Structure-from-Motion techniques. However, these approaches are too resource intensive to run in real-time, let alone to be implemented on mobile devices. In this paper, we propose to combine the scalability of such a global localization system running on a server with the speed and precision of a local pose tracker on a mobile device. Our approach is both scalable and drift-free by design and eliminates the need for loop closure. We propose two strategies to combine the information provided by local tracking and global localization. We evaluate our system on a large-scale dataset of the historic inner city of Aachen where it achieves interactive framerates at a localization error of less than 50cm while using less than 5MB of memory on the mobile device.

Photo tourism: exploring photo collections in 3D We present a system for interactively browsing and exploring large unstructured collections of photographs of a scene using a novel 3D interface. Our system consists of an image-based modeling front end that automatically computes the viewpoint of each photograph as well as a sparse 3D model of the scene and image to model correspondences. Our photo explorer uses image-based rendering techniques to smoothly transition between photographs, while also enabling full 3D navigation and exploration of the set of images and world geometry, along with auxiliary information such as overhead maps. Our system also makes it easy to construct photo tours of scenic or historic locations, and to annotate image details, which are automatically transferred to other relevant images. We demonstrate our system on several large personal photo collections as well as images gathered from Internet photo sharing sites.

On the security of public key protocols Recently the use of public key encryption to provide secure network communication has received considerable attention. Such public key systems are usually effective against passive eavesdroppers, who merely tap the lines and try to decipher the message. It has been pointed out, however, that an improperly designed protocol could be vulnerable to an active saboteur, one who may impersonate another user or alter the message being transmitted. Several models are formulated in which the security of protocols can be discussed precisely. Algorithms and characterizations that can be used to determine protocol security in these models are given.

Evolutionary computation: comments on the history and current state Evolutionary computation has started to receive significant attention during the last decade, although the origins can be traced back to the late 1950's. This article surveys the history as well as the current state of this rapidly growing field. We describe the purpose, the general structure, and the working principles of different approaches, including genetic algorithms (GA) (with links to genetic programming (GP) and classifier systems (CS)), evolution strategies (ES), and evolutionary programming (EP) by analysis and comparison of their most important constituents (i.e. representations, variation operators, reproduction, and selection mechanism). Finally, we give a brief overview on the manifold of application domains, although this necessarily must remain incomplete

Social navigation: techniques for building more usable systems

A study on the use of non-parametric tests for analyzing the evolutionary algorithms' behaviour: a case study on the CEC'2005 Special Session on Real Parameter Optimization In recent years, there has been a growing interest for the experimental analysis in the field of evolutionary algorithms. It is noticeable due to the existence of numerous papers which analyze and propose different types of problems, such as the basis for experimental comparisons of algorithms, proposals of different methodologies in comparison or proposals of use of different statistical techniques in algorithms’ comparison.In this paper, we focus our study on the use of statistical techniques in the analysis of evolutionary algorithms’ behaviour over optimization problems. A study about the required conditions for statistical analysis of the results is presented by using some models of evolutionary algorithms for real-coding optimization. This study is conducted in two ways: single-problem analysis and multiple-problem analysis. The results obtained state that a parametric statistical analysis could not be appropriate specially when we deal with multiple-problem results. In multiple-problem analysis, we propose the use of non-parametric statistical tests given that they are less restrictive than parametric ones and they can be used over small size samples of results. As a case study, we analyze the published results for the algorithms presented in the CEC’2005 Special Session on Real Parameter Optimization by using non-parametric test procedures.

Empirical Modelling of Genetic Algorithms This paper addresses the problem of reliably setting genetic algorithm parameters for consistent labelling problems. Genetic algorithm parameters are notoriously difficult to determine. This paper proposes a robust empirical framework, based on the analysis of factorial experiments. The use of a graeco-latin square permits an initial study of a wide range of parameter settings. This is followed by fully crossed factorial experiments with narrower ranges, which allow detailed analysis by logistic regression. The empirical models derived can be used to determine optimal algorithm parameters and to shed light on interactions between the parameters and their relative importance. Re-fined models are produced, which are shown to be robust under extrapolation to up to triple the problem size.

GROPING: Geomagnetism and cROwdsensing Powered Indoor NaviGation Although a large number of WiFi fingerprinting based indoor localization systems have been proposed, our field experience with Google Maps Indoor (GMI), the only system available for public testing, shows that it is far from mature for indoor navigation. In this paper, we first report our field studies with GMI, as well as experiment results aiming to explain our unsatisfactory GMI experience. Then motivated by the obtained insights, we propose GROPING as a self-contained indoor navigation system independent of any infrastructural support. GROPING relies on geomagnetic fingerprints that are far more stable than WiFi fingerprints, and it exploits crowdsensing to construct floor maps rather than expecting individual venues to supply digitized maps. Based on our experiments with 20 participants in various floors of a big shopping mall, GROPING is able to deliver a sufficient accuracy for localization and thus provides smooth navigation experience.

Inter-class sparsity based discriminative least square regression Least square regression is a very popular supervised classification method. However, two main issues greatly limit its performance. The first one is that it only focuses on fitting the input features to the corresponding output labels while ignoring the correlations among samples. The second one is that the used label matrix, i.e., zero–one label matrix is inappropriate for classification. To solve these problems and improve the performance, this paper presents a novel method, i.e., inter-class sparsity based discriminative least square regression (ICS_DLSR), for multi-class classification. Different from other methods, the proposed method pursues that the transformed samples have a common sparsity structure in each class. For this goal, an inter-class sparsity constraint is introduced to the least square regression model such that the margins of samples from the same class can be greatly reduced while those of samples from different classes can be enlarged. In addition, an error term with row-sparsity constraint is introduced to relax the strict zero–one label matrix, which allows the method to be more flexible in learning the discriminative transformation matrix. These factors encourage the method to learn a more compact and discriminative transformation for regression and thus has the potential to perform better than other methods. Extensive experimental results show that the proposed method achieves the best performance in comparison with other methods for multi-class classification.

Hardware Circuits Design and Performance Evaluation of a Soft Lower Limb Exoskeleton Soft lower limb exoskeletons (LLEs) are wearable devices that have good potential in walking rehabilitation and augmentation. While a few studies focused on the structure design and assistance force optimization of the soft LLEs, rarely work has been conducted on the hardware circuits design. The main purpose of this work is to present a new soft LLE for walking efficiency improvement and introduce its hardware circuits design. A soft LLE for hip flexion assistance and a hardware circuits system with scalability were proposed. To assess the efficacy of the soft LLE, the experimental tests that evaluate the sensor data acquisition, force tracking performance, lower limb muscle activity and metabolic cost were conducted. The time error in the peak assistance force was just 1%. The reduction in the normalized root-mean-square EMG of the rectus femoris was 7.1%. The net metabolic cost in exoskeleton on condition was reduced by 7.8% relative to walking with no exoskeleton. The results show that the designed hardware circuits can be applied to the soft LLE and the soft LLE is able to improve walking efficiency of wearers.

Rotated Sphere Haar Wavelet and Deep Contractive Auto-Encoder Network With Fuzzy Gaussian SVM for Pilot’s Pupil Center Detection How to track the attention of the pilot is a huge challenge. We are able to capture the pupil status of the pilot and analyze their anomalies and judge the attention of the pilot. This paper proposes a new approach to solve this problem through the integration of spherical Haar wavelet transform and deep learning methods. First, considering the application limitations of Haar wavelet and other wavelets in spherical signal decomposition and reconstruction, a feature learning method based on the spherical Haar wavelet is proposed. In order to obtain the salient features of the spherical signal, a rotating spherical Haar wavelet is also proposed, which has a consistent scale in the same direction between the reconstructed image and the original image. Second, in order to find a better characteristic representation of the spherical signal, a higher contractive autoencoder (HCAE) is designed for the potential representation of the spherical Haar wavelet coefficients, which has two penalty items, respectively, from Jacobian and two order items from Taylor expansion of the point ${x}$ for the contract learning of sample space. Third, in order to improve the classification performance, this paper proposes a fuzzy Gaussian support vector machine (FGSVM) as the top classification tool of the deep learning model, which can punish some Gaussian noise from the output of the deep HCAE network (DHCAEN). Finally, a DHCAEN-FGSVM classifier is proposed to identify the location of the pupil center. The experimental results of the public data set and actual data show that our model is an effective method for spherical signal detection.

Adaptive impedance control of a robotic orthosis for gait rehabilitation. Intervention of robotic devices in the field of physical gait therapy can help in providing repetitive, systematic, and economically viable training sessions. Interactive or assist-as-needed (AAN) gait training encourages patient voluntary participation in the robotic gait training process which may aid in rapid motor function recovery. In this paper, a lightweight robotic gait training orthosis w...

Hierarchical Compliance Control of a Soft Ankle Rehabilitation Robot Actuated by Pneumatic Muscles. Traditional compliance control of a rehabilitation robot is implemented in task space by using impedance or admittance control algorithms. The soft robot actuated by pneumatic muscle actuators (PMAs) is becoming prominent for patients as it enables the compliance being adjusted in each active link, which, however, has not been reported in the literature. This paper proposes a new compliance control method of a soft ankle rehabilitation robot that is driven by four PMAs configured in parallel to enable three degrees of freedom movement of the ankle joint. A new hierarchical compliance control structure, including a low-level compliance adjustment controller in joint space and a high-level admittance controller in task space, is designed. An adaptive compliance control paradigm is further developed by taking into account patient's active contribution and movement ability during a previous period of time, in order to provide robot assistance only when it is necessarily required. Experiments on healthy and impaired human subjects were conducted to verify the adaptive hierarchical compliance control scheme. The results show that the robot hierarchical compliance can be online adjusted according to the participant's assessment. The robot reduces its assistance output when participants contribute more and vice versa, thus providing a potentially feasible solution to the patient-in-loop cooperative training strategy.

An <italic>L</italic>₁-and-<italic>L</italic>₂-Norm-Oriented Latent Factor Model for Recommender Systems A recommender system (RS) is highly efficient in filtering people’s desired information from high-dimensional and sparse (HiDS) data. To date, a latent factor (LF)-based approach becomes highly popular when implementing a RS. However, current LF models mostly adopt single distance-oriented Loss like an <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$L_{2}$ </tex-math></inline-formula> norm-oriented one, which ignores target data’s characteristics described by other metrics like an <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$L_{1}$ </tex-math></inline-formula> norm-oriented one. To investigate this issue, this article proposes an <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$L_{1}$ </tex-math></inline-formula> -and- <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$L_{2}$ </tex-math></inline-formula> -norm-oriented LF ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\text{L}^{3}\text{F}$ </tex-math></inline-formula> ) model. It adopts twofold ideas: 1) aggregating <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$L_{1}$ </tex-math></inline-formula> norm’s robustness and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$L_{2}$ </tex-math></inline-formula> norm’s stability to form its Loss and 2) adaptively adjusting weights of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$L_{1}$ </tex-math></inline-formula> and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$L_{2}$ </tex-math></inline-formula> norms in its Loss. By doing so, it achieves fine aggregation effects with <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$L_{1}$ </tex-math></inline-formula> norm-oriented Loss ’s robustness and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$L_{2}$ </tex-math></inline-formula> norm-oriented Loss ’s stability to precisely describe HiDS data with outliers. Experimental results on nine HiDS datasets generated by real systems show that an <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\text{L}^{3}\text{F}$ </tex-math></inline-formula> model significantly outperforms state-of-the-art models in prediction accuracy for missing data of an HiDS dataset. Its computational efficiency is also comparable with the most efficient LF models. Hence, it has good potential for addressing HiDS data from real applications.

Evolutionary multi-objective automatic clustering enhanced with quality metrics and ensemble strategy Automatic clustering problem, which needs to detect the appropriate clustering without a pre-defined number of clusters (k), is difficult and challenging in unsupervised learning owing to the lack of prior domain knowledge. Despite a rising tendency with the application of evolutionary multi-objective optimization (EMO) techniques for automatic clustering, there still exist some obvious under-explored issues. In this paper, we resort to quality metrics and ensemble strategy for the sake of explicit/implicit knowledge discovery to guide the optimization process. The quality and diversity of solutions defined in terms of cluster validities, as similar to performance indicator for multi-objective optimization, are applied to assist in addressing automatic clustering problems and decreasing unnecessary computational overhead. To be specific, the main components like initialization, reproduction operations, and environmental selection which involved during EMO based automatic clustering are discussed and refined. For the determination of the final partitioning, quality metrics and cluster ensemble strategy are both considered to improve the retrieve system in the unsupervised way. Experiments are conducted from several different aspects and the corresponding analyses are provided, which confirm that the proposals are more efficient and effective for automatic clustering.

Evolutionary Multitasking via Explicit Autoencoding. Evolutionary multitasking (EMT) is an emerging research topic in the field of evolutionary computation. In contrast to the traditional single-task evolutionary search, EMT conducts evolutionary search on multiple tasks simultaneously. It aims to improve convergence characteristics across multiple optimization problems at once by seamlessly transferring knowledge among them. Due to the efficacy of EMT, it has attracted lots of research attentions and several EMT algorithms have been proposed in the literature. However, existing EMT algorithms are usually based on a common mode of knowledge transfer in the form of implicit genetic transfer through chromosomal crossover. This mode cannot make use of multiple biases embedded in different evolutionary search operators, which could give better search performance when properly harnessed. Keeping this in mind, this paper proposes an EMT algorithm with explicit genetic transfer across tasks, namely EMT via autoencoding, which allows the incorporation of multiple search mechanisms with different biases in the EMT paradigm. To confirm the efficacy of the proposed EMT algorithm with explicit autoencoding, comprehensive empirical studies have been conducted on both the single- and multi-objective multitask optimization problems.

Comprehensive Learning Particle Swarm Optimization Algorithm with Local Search for Multimodal Functions A comprehensive learning particle swarm optimizer (CLPSO) embedded with local search (LS) is proposed to pursue higher optimization performance by taking the advantages of CLPSO’s strong global search capability and LS’s fast convergence ability. This work proposes an adaptive LS starting strategy by utilizing our proposed quasi-entropy index to address its key issue, i.e., when to start LS. The changes of the index as the optimization proceeds are analyzed in theory and via numerical tests. The proposed algorithm is tested on multimodal benchmark functions. Parameter sensitivity analysis is performed to demonstrate its robustness. The comparison results reveal overall higher convergence rate and accuracy than those of CLPSO, state-of-the-art PSO variants.

Deep learning Deep learning allows computational models that are composed of multiple processing layers to learn representations of data with multiple levels of abstraction. These methods have dramatically improved the state-of-the-art in speech recognition, visual object recognition, object detection and many other domains such as drug discovery and genomics. Deep learning discovers intricate structure in large data sets by using the backpropagation algorithm to indicate how a machine should change its internal parameters that are used to compute the representation in each layer from the representation in the previous layer. Deep convolutional nets have brought about breakthroughs in processing images, video, speech and audio, whereas recurrent nets have shone light on sequential data such as text and speech. Machine-learning technology powers many aspects of modern society: from web searches to content filtering on social networks to recommendations on e-commerce websites, and it is increasingly present in consumer products such as cameras and smartphones. Machine-learning systems are used to identify objects in images, transcribe speech into text, match news items, posts or products with users' interests, and select relevant results of search. Increasingly, these applications make use of a class of techniques called deep learning. Conventional machine-learning techniques were limited in their ability to process natural data in their raw form. For decades, constructing a pattern-recognition or machine-learning system required careful engineering and considerable domain expertise to design a feature extractor that transformed the raw data (such as the pixel values of an image) into a suitable internal representation or feature vector from which the learning subsystem, often a classifier, could detect or classify patterns in the input. Representation learning is a set of methods that allows a machine to be fed with raw data and to automatically discover the representations needed for detection or classification. Deep-learning methods are representation-learning methods with multiple levels of representation, obtained by composing simple but non-linear modules that each transform the representation at one level (starting with the raw input) into a representation at a higher, slightly more abstract level. With the composition of enough such transformations, very complex functions can be learned. For classification tasks, higher layers of representation amplify aspects of the input that are important for discrimination and suppress irrelevant variations. An image, for example, comes in the form of an array of pixel values, and the learned features in the first layer of representation typically represent the presence or absence of edges at particular orientations and locations in the image. The second layer typically detects motifs by spotting particular arrangements of edges, regardless of small variations in the edge positions. The third layer may assemble motifs into larger combinations that correspond to parts of familiar objects, and subsequent layers would detect objects as combinations of these parts. The key aspect of deep learning is that these layers of features are not designed by human engineers: they are learned from data using a general-purpose learning procedure. Deep learning is making major advances in solving problems that have resisted the best attempts of the artificial intelligence community for many years. It has turned out to be very good at discovering intricate structures in high-dimensional data and is therefore applicable to many domains of science, business and government. In addition to beating records in image recognition1, 2, 3, 4 and speech recognition5, 6, 7, it has beaten other machine-learning techniques at predicting the activity of potential drug molecules8, analysing particle accelerator data9, 10, reconstructing brain circuits11, and predicting the effects of mutations in non-coding DNA on gene expression and disease12, 13. Perhaps more surprisingly, deep learning has produced extremely promising results for various tasks in natural language understanding14, particularly topic classification, sentiment analysis, question answering15 and language translation16, 17. We think that deep learning will have many more successes in the near future because it requires very little engineering by hand, so it can easily take advantage of increases in the amount of available computation and data. New learning algorithms and architectures that are currently being developed for deep neural networks will only accelerate this progress. The most common form of machine learning, deep or not, is supervised learning. Imagine that we want to build a system that can classify images as containing, say, a house, a car, a person or a pet. We first collect a large data set of images of houses, cars, people and pets, each labelled with its category. During training, the machine is shown an image and produces an output in the form of a vector of scores, one for each category. We want the desired category to have the highest score of all categories, but this is unlikely to happen before training. We compute an objective function that measures the error (or distance) between the output scores and the desired pattern of scores. The machine then modifies its internal adjustable parameters to reduce this error. These adjustable parameters, often called weights, are real numbers that can be seen as 'knobs' that define the input–output function of the machine. In a typical deep-learning system, there may be hundreds of millions of these adjustable weights, and hundreds of millions of labelled examples with which to train the machine. To properly adjust the weight vector, the learning algorithm computes a gradient vector that, for each weight, indicates by what amount the error would increase or decrease if the weight were increased by a tiny amount. The weight vector is then adjusted in the opposite direction to the gradient vector. The objective function, averaged over all the training examples, can be seen as a kind of hilly landscape in the high-dimensional space of weight values. The negative gradient vector indicates the direction of steepest descent in this landscape, taking it closer to a minimum, where the output error is low on average. In practice, most practitioners use a procedure called stochastic gradient descent (SGD). This consists of showing the input vector for a few examples, computing the outputs and the errors, computing the average gradient for those examples, and adjusting the weights accordingly. The process is repeated for many small sets of examples from the training set until the average of the objective function stops decreasing. It is called stochastic because each small set of examples gives a noisy estimate of the average gradient over all examples. This simple procedure usually finds a good set of weights surprisingly quickly when compared with far more elaborate optimization techniques18. After training, the performance of the system is measured on a different set of examples called a test set. This serves to test the generalization ability of the machine — its ability to produce sensible answers on new inputs that it has never seen during training. Many of the current practical applications of machine learning use linear classifiers on top of hand-engineered features. A two-class linear classifier computes a weighted sum of the feature vector components. If the weighted sum is above a threshold, the input is classified as belonging to a particular category. Since the 1960s we have known that linear classifiers can only carve their input space into very simple regions, namely half-spaces separated by a hyperplane19. But problems such as image and speech recognition require the input–output function to be insensitive to irrelevant variations of the input, such as variations in position, orientation or illumination of an object, or variations in the pitch or accent of speech, while being very sensitive to particular minute variations (for example, the difference between a white wolf and a breed of wolf-like white dog called a Samoyed). At the pixel level, images of two Samoyeds in different poses and in different environments may be very different from each other, whereas two images of a Samoyed and a wolf in the same position and on similar backgrounds may be very similar to each other. A linear classifier, or any other 'shallow' classifier operating on raw pixels could not possibly distinguish the latter two, while putting the former two in the same category. This is why shallow classifiers require a good feature extractor that solves the selectivity–invariance dilemma — one that produces representations that are selective to the aspects of the image that are important for discrimination, but that are invariant to irrelevant aspects such as the pose of the animal. To make classifiers more powerful, one can use generic non-linear features, as with kernel methods20, but generic features such as those arising with the Gaussian kernel do not allow the learner to generalize well far from the training examples21. The conventional option is to hand design good feature extractors, which requires a considerable amount of engineering skill and domain expertise. But this can all be avoided if good features can be learned automatically using a general-purpose learning procedure. This is the key advantage of deep learning. A deep-learning architecture is a multilayer stack of simple modules, all (or most) of which are subject to learning, and many of which compute non-linear input–output mappings. Each module in the stack transforms its input to increase both the selectivity and the invariance of the representation. With multiple non-linear layers, say a depth of 5 to 20, a system can implement extremely intricate functions of its inputs that are simultaneously sensitive to minute details — distinguishing Samoyeds from white wolves — and insensitive to large irrelevant variations such as the background, pose, lighting and surrounding objects. From the earliest days of pattern recognition22, 23, the aim of researchers has been to replace hand-engineered features with trainable multilayer networks, but despite its simplicity, the solution was not widely understood until the mid 1980s. As it turns out, multilayer architectures can be trained by simple stochastic gradient descent. As long as the modules are relatively smooth functions of their inputs and of their internal weights, one can compute gradients using the backpropagation procedure. The idea that this could be done, and that it worked, was discovered independently by several different groups during the 1970s and 1980s24, 25, 26, 27. The backpropagation procedure to compute the gradient of an objective function with respect to the weights of a multilayer stack of modules is nothing more than a practical application of the chain rule for derivatives. The key insight is that the derivative (or gradient) of the objective with respect to the input of a module can be computed by working backwards from the gradient with respect to the output of that module (or the input of the subsequent module) (Fig. 1). The backpropagation equation can be applied repeatedly to propagate gradients through all modules, starting from the output at the top (where the network produces its prediction) all the way to the bottom (where the external input is fed). Once these gradients have been computed, it is straightforward to compute the gradients with respect to the weights of each module. Many applications of deep learning use feedforward neural network architectures (Fig. 1), which learn to map a fixed-size input (for example, an image) to a fixed-size output (for example, a probability for each of several categories). To go from one layer to the next, a set of units compute a weighted sum of their inputs from the previous layer and pass the result through a non-linear function. At present, the most popular non-linear function is the rectified linear unit (ReLU), which is simply the half-wave rectifier f(z) = max(z, 0). In past decades, neural nets used smoother non-linearities, such as tanh(z) or 1/(1 + exp(−z)), but the ReLU typically learns much faster in networks with many layers, allowing training of a deep supervised network without unsupervised pre-training28. Units that are not in the input or output layer are conventionally called hidden units. The hidden layers can be seen as distorting the input in a non-linear way so that categories become linearly separable by the last layer (Fig. 1). In the late 1990s, neural nets and backpropagation were largely forsaken by the machine-learning community and ignored by the computer-vision and speech-recognition communities. It was widely thought that learning useful, multistage, feature extractors with little prior knowledge was infeasible. In particular, it was commonly thought that simple gradient descent would get trapped in poor local minima — weight configurations for which no small change would reduce the average error. In practice, poor local minima are rarely a problem with large networks. Regardless of the initial conditions, the system nearly always reaches solutions of very similar quality. Recent theoretical and empirical results strongly suggest that local minima are not a serious issue in general. Instead, the landscape is packed with a combinatorially large number of saddle points where the gradient is zero, and the surface curves up in most dimensions and curves down in the remainder29, 30. The analysis seems to show that saddle points with only a few downward curving directions are present in very large numbers, but almost all of them have very similar values of the objective function. Hence, it does not much matter which of these saddle points the algorithm gets stuck at. Interest in deep feedforward networks was revived around 2006 (refs 31,32,33,34) by a group of researchers brought together by the Canadian Institute for Advanced Research (CIFAR). The researchers introduced unsupervised learning procedures that could create layers of feature detectors without requiring labelled data. The objective in learning each layer of feature detectors was to be able to reconstruct or model the activities of feature detectors (or raw inputs) in the layer below. By 'pre-training' several layers of progressively more complex feature detectors using this reconstruction objective, the weights of a deep network could be initialized to sensible values. A final layer of output units could then be added to the top of the network and the whole deep system could be fine-tuned using standard backpropagation33, 34, 35. This worked remarkably well for recognizing handwritten digits or for detecting pedestrians, especially when the amount of labelled data was very limited36. The first major application of this pre-training approach was in speech recognition, and it was made possible by the advent of fast graphics processing units (GPUs) that were convenient to program37 and allowed researchers to train networks 10 or 20 times faster. In 2009, the approach was used to map short temporal windows of coefficients extracted from a sound wave to a set of probabilities for the various fragments of speech that might be represented by the frame in the centre of the window. It achieved record-breaking results on a standard speech recognition benchmark that used a small vocabulary38 and was quickly developed to give record-breaking results on a large vocabulary task39. By 2012, versions of the deep net from 2009 were being developed by many of the major speech groups6 and were already being deployed in Android phones. For smaller data sets, unsupervised pre-training helps to prevent overfitting40, leading to significantly better generalization when the number of labelled examples is small, or in a transfer setting where we have lots of examples for some 'source' tasks but very few for some 'target' tasks. Once deep learning had been rehabilitated, it turned out that the pre-training stage was only needed for small data sets. There was, however, one particular type of deep, feedforward network that was much easier to train and generalized much better than networks with full connectivity between adjacent layers. This was the convolutional neural network (ConvNet)41, 42. It achieved many practical successes during the period when neural networks were out of favour and it has recently been widely adopted by the computer-vision community. ConvNets are designed to process data that come in the form of multiple arrays, for example a colour image composed of three 2D arrays containing pixel intensities in the three colour channels. Many data modalities are in the form of multiple arrays: 1D for signals and sequences, including language; 2D for images or audio spectrograms; and 3D for video or volumetric images. There are four key ideas behind ConvNets that take advantage of the properties of natural signals: local connections, shared weights, pooling and the use of many layers. The architecture of a typical ConvNet (Fig. 2) is structured as a series of stages. The first few stages are composed of two types of layers: convolutional layers and pooling layers. Units in a convolutional layer are organized in feature maps, within which each unit is connected to local patches in the feature maps of the previous layer through a set of weights called a filter bank. The result of this local weighted sum is then passed through a non-linearity such as a ReLU. All units in a feature map share the same filter bank. Different feature maps in a layer use different filter banks. The reason for this architecture is twofold. First, in array data such as images, local groups of values are often highly correlated, forming distinctive local motifs that are easily detected. Second, the local statistics of images and other signals are invariant to location. In other words, if a motif can appear in one part of the image, it could appear anywhere, hence the idea of units at different locations sharing the same weights and detecting the same pattern in different parts of the array. Mathematically, the filtering operation performed by a feature map is a discrete convolution, hence the name. Although the role of the convolutional layer is to detect local conjunctions of features from the previous layer, the role of the pooling layer is to merge semantically similar features into one. Because the relative positions of the features forming a motif can vary somewhat, reliably detecting the motif can be done by coarse-graining the position of each feature. A typical pooling unit computes the maximum of a local patch of units in one feature map (or in a few feature maps). Neighbouring pooling units take input from patches that are shifted by more than one row or column, thereby reducing the dimension of the representation and creating an invariance to small shifts and distortions. Two or three stages of convolution, non-linearity and pooling are stacked, followed by more convolutional and fully-connected layers. Backpropagating gradients through a ConvNet is as simple as through a regular deep network, allowing all the weights in all the filter banks to be trained. Deep neural networks exploit the property that many natural signals are compositional hierarchies, in which higher-level features are obtained by composing lower-level ones. In images, local combinations of edges form motifs, motifs assemble into parts, and parts form objects. Similar hierarchies exist in speech and text from sounds to phones, phonemes, syllables, words and sentences. The pooling allows representations to vary very little when elements in the previous layer vary in position and appearance. The convolutional and pooling layers in ConvNets are directly inspired by the classic notions of simple cells and complex cells in visual neuroscience43, and the overall architecture is reminiscent of the LGN–V1–V2–V4–IT hierarchy in the visual cortex ventral pathway44. When ConvNet models and monkeys are shown the same picture, the activations of high-level units in the ConvNet explains half of the variance of random sets of 160 neurons in the monkey's inferotemporal cortex45. ConvNets have their roots in the neocognitron46, the architecture of which was somewhat similar, but did not have an end-to-end supervised-learning algorithm such as backpropagation. A primitive 1D ConvNet called a time-delay neural net was used for the recognition of phonemes and simple words47, 48. There have been numerous applications of convolutional networks going back to the early 1990s, starting with time-delay neural networks for speech recognition47 and document reading42. The document reading system used a ConvNet trained jointly with a probabilistic model that implemented language constraints. By the late 1990s this system was reading over 10% of all the cheques in the United States. A number of ConvNet-based optical character recognition and handwriting recognition systems were later deployed by Microsoft49. ConvNets were also experimented with in the early 1990s for object detection in natural images, including faces and hands50, 51, and for face recognition52. Since the early 2000s, ConvNets have been applied with great success to the detection, segmentation and recognition of objects and regions in images. These were all tasks in which labelled data was relatively abundant, such as traffic sign recognition53, the segmentation of biological images54 particularly for connectomics55, and the detection of faces, text, pedestrians and human bodies in natural images36, 50, 51, 56, 57, 58. A major recent practical success of ConvNets is face recognition59. Importantly, images can be labelled at the pixel level, which will have applications in technology, including autonomous mobile robots and self-driving cars60, 61. Companies such as Mobileye and NVIDIA are using such ConvNet-based methods in their upcoming vision systems for cars. Other applications gaining importance involve natural language understanding14 and speech recognition7. Despite these successes, ConvNets were largely forsaken by the mainstream computer-vision and machine-learning communities until the ImageNet competition in 2012. When deep convolutional networks were applied to a data set of about a million images from the web that contained 1,000 different classes, they achieved spectacular results, almost halving the error rates of the best competing approaches1. This success came from the efficient use of GPUs, ReLUs, a new regularization technique called dropout62, and techniques to generate more training examples by deforming the existing ones. This success has brought about a revolution in computer vision; ConvNets are now the dominant approach for almost all recognition and detection tasks4, 58, 59, 63, 64, 65 and approach human performance on some tasks. A recent stunning demonstration combines ConvNets and recurrent net modules for the generation of image captions (Fig. 3). Recent ConvNet architectures have 10 to 20 layers of ReLUs, hundreds of millions of weights, and billions of connections between units. Whereas training such large networks could have taken weeks only two years ago, progress in hardware, software and algorithm parallelization have reduced training times to a few hours. The performance of ConvNet-based vision systems has caused most major technology companies, including Google, Facebook, Microsoft, IBM, Yahoo!, Twitter and Adobe, as well as a quickly growing number of start-ups to initiate research and development projects and to deploy ConvNet-based image understanding products and services. ConvNets are easily amenable to efficient hardware implementations in chips or field-programmable gate arrays66, 67. A number of companies such as NVIDIA, Mobileye, Intel, Qualcomm and Samsung are developing ConvNet chips to enable real-time vision applications in smartphones, cameras, robots and self-driving cars. Deep-learning theory shows that deep nets have two different exponential advantages over classic learning algorithms that do not use distributed representations21. Both of these advantages arise from the power of composition and depend on the underlying data-generating distribution having an appropriate componential structure40. First, learning distributed representations enable generalization to new combinations of the values of learned features beyond those seen during training (for example, 2n combinations are possible with n binary features)68, 69. Second, composing layers of representation in a deep net brings the potential for another exponential advantage70 (exponential in the depth). The hidden layers of a multilayer neural network learn to represent the network's inputs in a way that makes it easy to predict the target outputs. This is nicely demonstrated by training a multilayer neural network to predict the next word in a sequence from a local context of earlier words71. Each word in the context is presented to the network as a one-of-N vector, that is, one component has a value of 1 and the rest are 0. In the first layer, each word creates a different pattern of activations, or word vectors (Fig. 4). In a language model, the other layers of the network learn to convert the input word vectors into an output word vector for the predicted next word, which can be used to predict the probability for any word in the vocabulary to appear as the next word. The network learns word vectors that contain many active components each of which can be interpreted as a separate feature of the word, as was first demonstrated27 in the context of learning distributed representations for symbols. These semantic features were not explicitly present in the input. They were discovered by the learning procedure as a good way of factorizing the structured relationships between the input and output symbols into multiple 'micro-rules'. Learning word vectors turned out to also work very well when the word sequences come from a large corpus of real text and the individual micro-rules are unreliable71. When trained to predict the next word in a news story, for example, the learned word vectors for Tuesday and Wednesday are very similar, as are the word vectors for Sweden and Norway. Such representations are called distributed representations because their elements (the features) are not mutually exclusive and their many configurations correspond to the variations seen in the observed data. These word vectors are composed of learned features that were not determined ahead of time by experts, but automatically discovered by the neural network. Vector representations of words learned from text are now very widely used in natural language applications14, 17, 72, 73, 74, 75, 76. The issue of representation lies at the heart of the debate between the logic-inspired and the neural-network-inspired paradigms for cognition. In the logic-inspired paradigm, an instance of a symbol is something for which the only property is that it is either identical or non-identical to other symbol instances. It has no internal structure that is relevant to its use; and to reason with symbols, they must be bound to the variables in judiciously chosen rules of inference. By contrast, neural networks just use big activity vectors, big weight matrices and scalar non-linearities to perform the type of fast 'intuitive' inference that underpins effortless commonsense reasoning. Before the introduction of neural language models71, the standard approach to statistical modelling of language did not exploit distributed representations: it was based on counting frequencies of occurrences of short symbol sequences of length up to N (called N-grams). The number of possible N-grams is on the order of VN, where V is the vocabulary size, so taking into account a context of more than a handful of words would require very large training corpora. N-grams treat each word as an atomic unit, so they cannot generalize across semantically related sequences of words, whereas neural language models can because they associate each word with a vector of real valued features, and semantically related words end up close to each other in that vector space (Fig. 4). When backpropagation was first introduced, its most exciting use was for training recurrent neural networks (RNNs). For tasks that involve sequential inputs, such as speech and language, it is often better to use RNNs (Fig. 5). RNNs process an input sequence one element at a time, maintaining in their hidden units a 'state vector' that implicitly contains information about the history of all the past elements of the sequence. When we consider the outputs of the hidden units at different discrete time steps as if they were the outputs of different neurons in a deep multilayer network (Fig. 5, right), it becomes clear how we can apply backpropagation to train RNNs. RNNs are very powerful dynamic systems, but training them has proved to be problematic because the backpropagated gradients either grow or shrink at each time step, so over many time steps they typically explode or vanish77, 78. Thanks to advances in their architecture79, 80 and ways of training them81, 82, RNNs have been found to be very good at predicting the next character in the text83 or the next word in a sequence75, but they can also be used for more complex tasks. For example, after reading an English sentence one word at a time, an English 'encoder' network can be trained so that the final state vector of its hidden units is a good representation of the thought expressed by the sentence. This thought vector can then be used as the initial hidden state of (or as extra input to) a jointly trained French 'decoder' network, which outputs a probability distribution for the first word of the French translation. If a particular first word is chosen from this distribution and provided as input to the decoder network it will then output a probability distribution for the second word of the translation and so on until a full stop is chosen17, 72, 76. Overall, this process generates sequences of French words according to a probability distribution that depends on the English sentence. This rather naive way of performing machine translation has quickly become competitive with the state-of-the-art, and this raises serious doubts about whether understanding a sentence requires anything like the internal symbolic expressions that are manipulated by using inference rules. It is more compatible with the view that everyday reasoning involves many simultaneous analogies that each contribute plausibility to a conclusion84, 85. Instead of translating the meaning of a French sentence into an English sentence, one can learn to 'translate' the meaning of an image into an English sentence (Fig. 3). The encoder here is a deep ConvNet that converts the pixels into an activity vector in its last hidden layer. The decoder is an RNN similar to the ones used for machine translation and neural language modelling. There has been a surge of interest in such systems recently (see examples mentioned in ref. 86). RNNs, once unfolded in time (Fig. 5), can be seen as very deep feedforward networks in which all the layers share the same weights. Although their main purpose is to learn long-term dependencies, theoretical and empirical evidence shows that it is difficult to learn to store information for very long78. To correct for that, one idea is to augment the network with an explicit memory. The first proposal of this kind is the long short-term memory (LSTM) networks that use special hidden units, the natural behaviour of which is to remember inputs for a long time79. A special unit called the memory cell acts like an accumulator or a gated leaky neuron: it has a connection to itself at the next time step that has a weight of one, so it copies its own real-valued state and accumulates the external signal, but this self-connection is multiplicatively gated by another unit that learns to decide when to clear the content of the memory. LSTM networks have subsequently proved to be more effective than conventional RNNs, especially when they have several layers for each time step87, enabling an entire speech recognition system that goes all the way from acoustics to the sequence of characters in the transcription. LSTM networks or related forms of gated units are also currently used for the encoder and decoder networks that perform so well at machine translation17, 72, 76. Over the past year, several authors have made different proposals to augment RNNs with a memory module. Proposals include the Neural Turing Machine in which the network is augmented by a 'tape-like' memory that the RNN can choose to read from or write to88, and memory networks, in which a regular network is augmented by a kind of associative memory89. Memory networks have yielded excellent performance on standard question-answering benchmarks. The memory is used to remember the story about which the network is later asked to answer questions. Beyond simple memorization, neural Turing machines and memory networks are being used for tasks that would normally require reasoning and symbol manipulation. Neural Turing machines can be taught 'algorithms'. Among other things, they can learn to output a sorted list of symbols when their input consists of an unsorted sequence in which each symbol is accompanied by a real value that indicates its priority in the list88. Memory networks can be trained to keep track of the state of the world in a setting similar to a text adventure game and after reading a story, they can answer questions that require complex inference90. In one test example, the network is shown a 15-sentence version of the The Lord of the Rings and correctly answers questions such as “where is Frodo now?”89. Unsupervised learning91, 92, 93, 94, 95, 96, 97, 98 had a catalytic effect in reviving interest in deep learning, but has since been overshadowed by the successes of purely supervised learning. Although we have not focused on it in this Review, we expect unsupervised learning to become far more important in the longer term. Human and animal learning is largely unsupervised: we discover the structure of the world by observing it, not by being told the name of every object. Human vision is an active process that sequentially samples the optic array in an intelligent, task-specific way using a small, high-resolution fovea with a large, low-resolution surround. We expect much of the future progress in vision to come from systems that are trained end-to-end and combine ConvNets with RNNs that use reinforcement learning to decide where to look. Systems combining deep learning and reinforcement learning are in their infancy, but they already outperform passive vision systems99 at classification tasks and produce impressive results in learning to play many different video games100. Natural language understanding is another area in which deep learning is poised to make a large impact over the next few years. We expect systems that use RNNs to understand sentences or whole documents will become much better when they learn strategies for selectively attending to one part at a time76, 86. Ultimately, major progress in artificial intelligence will come about through systems that combine representation learning with complex reasoning. Although deep learning and simple reasoning have been used for speech and handwriting recognition for a long time, new paradigms are needed to replace rule-based manipulation of symbolic expressions by operations on large vectors101. Download references The authors would like to thank the Natural Sciences and Engineering Research Council of Canada, the Canadian Institute For Advanced Research (CIFAR), the National Science Foundation and Office of Naval Research for support. Y.L. and Y.B. are CIFAR fellows. Reprints and permissions information is available at www.nature.com/reprints.

A comparative study of texture measures with classification based on featured distributions This paper evaluates the performance both of some texture measures which have been successfully used in various applications and of some new promising approaches proposed recently. For classification a method based on Kullback discrimination of sample and prototype distributions is used. The classification results for single features with one-dimensional feature value distributions and for pairs of complementary features with two-dimensional distributions are presented

Response time in man-computer conversational transactions The literature concerning man-computer transactions abounds in controversy about the limits of "system response time" to a user's command or inquiry at a terminal. Two major semantic issues prohibit resolving this controversy. One issue centers around the question of "Response time to what?" The implication is that different human purposes and actions will have different acceptable or useful response times.

Adaptive Learning in Tracking Control Based on the Dual Critic Network Design. In this paper, we present a new adaptive dynamic programming approach by integrating a reference network that provides an internal goal representation to help the systems learning and optimization. Specifically, we build the reference network on top of the critic network to form a dual critic network design that contains the detailed internal goal representation to help approximate the value funct...

A novel data hiding for color images based on pixel value difference and modulus function This paper proposes a novel data hiding method using pixel-value difference and modulus function for color image with the large embedding capacity(hiding 810757 bits in a 512 512 host image at least) and a high-visual-quality of the cover image. The proposed method has fully taken into account the correlation of the R, G and B plane of a color image. The amount of information embedded the R plane and the B plane determined by the difference of the corresponding pixel value between the G plane and the median of G pixel value in each pixel block. Furthermore, two sophisticated pixel value adjustment processes are provided to maintain the division consistency and to solve underflow and overflow problems. The most importance is that the secret data are completely extracted through the mathematical theoretical proof.

Design of robust fuzzy fault detection filter for polynomial fuzzy systems with new finite frequency specifications This paper investigates the problem of fault detection filter design for discrete-time polynomial fuzzy systems with faults and unknown disturbances. The frequency ranges of the faults and the disturbances are assumed to be known beforehand and to reside in low, middle or high frequency ranges. Thus, the proposed filter is designed in the finite frequency range to overcome the conservatism generated by those designed in the full frequency domain. Being of polynomial fuzzy structure, the proposed filter combines the H−/H∞ performances in order to ensure the best robustness to the disturbance and the best sensitivity to the fault. Design conditions are derived in Sum Of Squares formulations that can be easily solved via available software tools. Two illustrative examples are introduced to demonstrate the effectiveness of the proposed method and a comparative study with LMI method is also provided.

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

Sleep-Scheduling-Based Hierarchical Data Collection Algorithm for Gliders in Underwater Acoustic Sensor Networks In recent years, underwater acoustic sensor networks (UASNs) have been widely investigated for ocean environmental monitoring, offshore exploration, and marine military. The core function of UASNs is to collect data for related operations. A number of factors make the monitoring challenging; ocean thermoclines may affect the communication of the underwater nodes and gliders, reducing their communi...

Energy- and Spectral-Efficiency Tradeoff for Distributed Antenna Systems with Proportional Fairness Energy efficiency(EE) has caught more and more attention in future wireless communications due to steadily rising energy costs and environmental concerns. In this paper, we propose an EE scheme with proportional fairness for the downlink multiuser distributed antenna systems (DAS). Our aim is to maximize EE, subject to constraints on overall transmit power of each remote access unit (RAU), bit-error rate (BER), and proportional data rates. We exploit multi-criteria optimization method to systematically investigate the relationship between EE and spectral efficiency (SE). Using the weighted sum method, we first convert the multi-criteria optimization problem, which is extremely complex, into a simpler single objective optimization problem. Then an optimal algorithm is developed to allocate the available power to balance the tradeoff between EE and SE. We also demonstrate the effectiveness of the proposed scheme and illustrate the fundamental tradeoff between energy- and spectral-efficient transmission through computer simulation.

Efficient multi-task allocation and path planning for unmanned surface vehicle in support of ocean operations. Presently, there is an increasing interest in the deployment of unmanned surface vehicles (USVs) to support complex ocean operations. In order to carry out these missions in a more efficient way, an intelligent hybrid multi-task allocation and path planning algorithm is required and has been proposed in this paper. In terms of the multi-task allocation, a novel algorithm based upon a self-organising map (SOM) has been designed and developed. The main contribution is that an adaptive artificial repulsive force field has been constructed and integrated into the SOM to achieve collision avoidance capability. The new algorithm is able to fast and effectively generate a sequence for executing multiple tasks in a cluttered maritime environment involving numerous obstacles. After generating an optimised task execution sequence, a path planning algorithm based upon fast marching square (FMS) is utilised to calculate the trajectories. Because of the introduction of a safety parameter, the FMS is able to adaptively adjust the dimensional influence of an obstacle and accordingly generate the paths to ensure the safety of the USV. The algorithms have been verified and evaluated through a number of computer based simulations and has been proven to work effectively in both simulated and practical maritime environments. (C) 2017 Elsevier B.V. All rights reserved.

Cooperative Internet of UAVs: Distributed Trajectory Design by Multi-Agent Deep Reinforcement Learning Due to the advantages of flexible deployment and extensive coverage, unmanned aerial vehicles (UAVs) have significant potential for sensing applications in the next generation of cellular networks, which will give rise to a cellular Internet of UAVs. In this article, we consider a cellular Internet of UAVs, where the UAVs execute sensing tasks through cooperative sensing and transmission to minimize the age of information (AoI). However, the cooperative sensing and transmission is tightly coupled with the UAVs' trajectories, which makes the trajectory design challenging. To tackle this challenge, we propose a distributed sense-and-send protocol, where the UAVs determine the trajectories by selecting from a discrete set of tasks and a continuous set of locations for sensing and transmission. Based on this protocol, we formulate the trajectory design problem for AoI minimization and propose a compound-action actor-critic (CA2C) algorithm to solve it based on deep reinforcement learning. The CA2C algorithm can learn the optimal policies for actions involving both continuous and discrete variables and is suited for the trajectory design. Our simulation results show that the CA2C algorithm outperforms four baseline algorithms. Also, we show that by dividing the tasks, cooperative UAVs can achieve a lower AoI compared to non-cooperative UAVs.

Disturbance Compensating Model Predictive Control With Application to Ship Heading Control. To address the constraint violation and feasibility issues of model predictive control (MPC) for ship heading control in wave fields, a novel disturbance compensating MPC (DC-MPC) algorithm has been proposed to satisfy the state constraints in the presence of environmental disturbances. The capability of the novel DC-MPC algorithm is first analyzed. Then, the proposed DC-MPC algorithm is applied to solve the ship heading control problem, and its performance is compared with a modified MPC controller, which considers the estimated disturbance in the optimization directly. The simulation results show good performance of the proposed controller in terms of reducing heading error and satisfying yaw velocity and actuator saturation constraints. The DC-MPC algorithm has the potential to be applied to other motion control problems with environmental disturbances, such as flight, automobile, and robotics control.

MagicNet: The Maritime Giant Cellular Network Recently, the development of marine industries has increasingly attracted attention from all over the world. A wide-area and seamless maritime communication network has become a critical supporting approach. In this article, we propose a novel architecture named the maritime giant cellular network (MagicNet) relying on seaborne floating towers deployed in a honeycomb topology. The tower-borne gian...

Hamming Embedding and Weak Geometric Consistency for Large Scale Image Search This paper improves recent methods for large scale image search. State-of-the-art methods build on the bag-of-features image representation. We, first, analyze bag-of-features in the framework of approximate nearest neighbor search. This shows the sub-optimality of such a representation for matching descriptors and leads us to derive a more precise representation based on 1) Hamming embedding (HE) and 2) weak geometric consistency constraints (WGC). HE provides binary signatures that refine the matching based on visual words. WGC filters matching descriptors that are not consistent in terms of angle and scale. HE and WGC are integrated within the inverted file and are efficiently exploited for all images, even in the case of very large datasets. Experiments performed on a dataset of one million of images show a significant improvement due to the binary signature and the weak geometric consistency constraints, as well as their efficiency. Estimation of the full geometric transformation, i.e., a re-ranking step on a short list of images, is complementary to our weak geometric consistency constraints and allows to further improve the accuracy.

Microsoft Coco: Common Objects In Context We present a new dataset with the goal of advancing the state-of-the-art in object recognition by placing the question of object recognition in the context of the broader question of scene understanding. This is achieved by gathering images of complex everyday scenes containing common objects in their natural context. Objects are labeled using per-instance segmentations to aid in precise object localization. Our dataset contains photos of 91 objects types that would be easily recognizable by a 4 year old. With a total of 2.5 million labeled instances in 328k images, the creation of our dataset drew upon extensive crowd worker involvement via novel user interfaces for category detection, instance spotting and instance segmentation. We present a detailed statistical analysis of the dataset in comparison to PASCAL, ImageNet, and SUN. Finally, we provide baseline performance analysis for bounding box and segmentation detection results using a Deformable Parts Model.

The Whale Optimization Algorithm. The Whale Optimization Algorithm inspired by humpback whales is proposed.The WOA algorithm is benchmarked on 29 well-known test functions.The results on the unimodal functions show the superior exploitation of WOA.The exploration ability of WOA is confirmed by the results on multimodal functions.The results on structural design problems confirm the performance of WOA in practice. This paper proposes a novel nature-inspired meta-heuristic optimization algorithm, called Whale Optimization Algorithm (WOA), which mimics the social behavior of humpback whales. The algorithm is inspired by the bubble-net hunting strategy. WOA is tested with 29 mathematical optimization problems and 6 structural design problems. Optimization results prove that the WOA algorithm is very competitive compared to the state-of-art meta-heuristic algorithms as well as conventional methods. The source codes of the WOA algorithm are publicly available at http://www.alimirjalili.com/WOA.html

Collaborative privacy management The landscape of the World Wide Web with all its versatile services heavily relies on the disclosure of private user information. Unfortunately, the growing amount of personal data collected by service providers poses a significant privacy threat for Internet users. Targeting growing privacy concerns of users, privacy-enhancing technologies emerged. One goal of these technologies is the provision of tools that facilitate a more informative decision about personal data disclosures. A famous PET representative is the PRIME project that aims for a holistic privacy-enhancing identity management system. However, approaches like the PRIME privacy architecture require service providers to change their server infrastructure and add specific privacy-enhancing components. In the near future, service providers are not expected to alter internal processes. Addressing the dependency on service providers, this paper introduces a user-centric privacy architecture that enables the provider-independent protection of personal data. A central component of the proposed privacy infrastructure is an online privacy community, which facilitates the open exchange of privacy-related information about service providers. We characterize the benefits and the potentials of our proposed solution and evaluate a prototypical implementation.

Data-Driven Intelligent Transportation Systems: A Survey For the last two decades, intelligent transportation systems (ITS) have emerged as an efficient way of improving the performance of transportation systems, enhancing travel security, and providing more choices to travelers. A significant change in ITS in recent years is that much more data are collected from a variety of sources and can be processed into various forms for different stakeholders. The availability of a large amount of data can potentially lead to a revolution in ITS development, changing an ITS from a conventional technology-driven system into a more powerful multifunctional data-driven intelligent transportation system (D2ITS) : a system that is vision, multisource, and learning algorithm driven to optimize its performance. Furthermore, D2ITS is trending to become a privacy-aware people-centric more intelligent system. In this paper, we provide a survey on the development of D2ITS, discussing the functionality of its key components and some deployment issues associated with D2ITS Future research directions for the development of D2ITS is also presented.

Completely Pinpointing the Missing RFID Tags in a Time-Efficient Way Radio Frequency Identification (RFID) technology has been widely used in inventory management in many scenarios, e.g., warehouses, retail stores, hospitals, etc. This paper investigates a challenging problem of complete identification of missing tags in large-scale RFID systems. Although this problem has attracted extensive attention from academy and industry, the existing work can hardly satisfy the stringent real-time requirements. In this paper, a Slot Filter-based Missing Tag Identification (SFMTI) protocol is proposed to reconcile some expected collision slots into singleton slots and filter out the expected empty slots as well as the unreconcilable collision slots, thereby achieving the improved time-efficiency. The theoretical analysis is conducted to minimize the execution time of the proposed SFMTI. We then propose a cost-effective method to extend SFMTI to the multi-reader scenarios. The extensive simulation experiments and performance results demonstrate that the proposed SFMTI protocol outperforms the most promising Iterative ID-free Protocol (IIP) by reducing nearly 45% of the required execution time, and is just within a factor of 1.18 from the lower bound of the minimum execution time.

Modeling taxi driver anticipatory behavior. As part of a wider behavioral agent-based model that simulates taxi drivers' dynamic passenger-finding behavior under uncertainty, we present a model of strategic behavior of taxi drivers in anticipation of substantial time varying demand at locations such as airports and major train stations. The model assumes that, considering a particular decision horizon, a taxi driver decides to transfer to such a destination based on a reward function. The dynamic uncertainty of demand is captured by a time dependent pick-up probability, which is a cumulative distribution function of waiting time. The model allows for information learning by which taxi drivers update their beliefs from past experiences. A simulation on a real road network, applied to test the model, indicates that the formulated model dynamically improves passenger-finding strategies at the airport. Taxi drivers learn when to transfer to the airport in anticipation of the time-varying demand at the airport to minimize their waiting time.

Hardware Circuits Design and Performance Evaluation of a Soft Lower Limb Exoskeleton Soft lower limb exoskeletons (LLEs) are wearable devices that have good potential in walking rehabilitation and augmentation. While a few studies focused on the structure design and assistance force optimization of the soft LLEs, rarely work has been conducted on the hardware circuits design. The main purpose of this work is to present a new soft LLE for walking efficiency improvement and introduce its hardware circuits design. A soft LLE for hip flexion assistance and a hardware circuits system with scalability were proposed. To assess the efficacy of the soft LLE, the experimental tests that evaluate the sensor data acquisition, force tracking performance, lower limb muscle activity and metabolic cost were conducted. The time error in the peak assistance force was just 1%. The reduction in the normalized root-mean-square EMG of the rectus femoris was 7.1%. The net metabolic cost in exoskeleton on condition was reduced by 7.8% relative to walking with no exoskeleton. The results show that the designed hardware circuits can be applied to the soft LLE and the soft LLE is able to improve walking efficiency of wearers.

Progressive Growing of GANs for Improved Quality, Stability, and Variation. We describe a new training methodology for generative adversarial networks. The key idea is to grow both the generator and discriminator progressively: starting from a low resolution, we add new layers that model increasingly fine details as training progresses. This both speeds the training up and greatly stabilizes it, allowing us to produce images of unprecedented quality, e.g., CelebA images at 1024^2. We also propose a simple way to increase the variation in generated images, and achieve a record inception score of 8.80 in unsupervised CIFAR10. Additionally, we describe several implementation details that are important for discouraging unhealthy competition between the generator and discriminator. Finally, we suggest a new metric for evaluating GAN results, both in terms of image quality and variation. As an additional contribution, we construct a higher-quality version of the CelebA dataset.

Mask-Guided Portrait Editing With Conditional Gans Portrait editing is a popular subject in photo manipulation. The Generative Adversarial Network (GAN) advances the generating of realistic faces and allows more face editing. In this paper, we argue about three issues in existing techniques: diversity, quality, and controllability for portrait synthesis and editing. To address these issues, we propose a novel end-to-end learning framework that leverages conditional GANs guided by provided face masks for generating faces. The framework learns feature embeddings for every face component (e.g., mouth, hair, eye), separately, contributing to better correspondences for image translation, and local face editing. With the mask, our network is available to many applications, like face synthesis driven by mask, face Swap+ (including hair in swapping), and local manipulation. It can also boost the performance of face parsing a bit as an option of data augmentation.

Improved Training of Wasserstein GANs. Generative Adversarial Networks (GANs) are powerful generative models, but suffer from training instability. The recently proposed Wasserstein GAN (WGAN) makes progress toward stable training of GANs, but sometimes can still generate only poor samples or fail to converge. We find that these problems are often due to the use of weight clipping in WGAN to enforce a Lipschitz constraint on the critic, which can lead to undesired behavior. We propose an alternative to clipping weights: penalize the norm of gradient of the critic with respect to its input. Our proposed method performs better than standard WGAN and enables stable training of a wide variety of GAN architectures with almost no hyperparameter tuning, including 101-layer ResNets and language models with continuous generators. We also achieve high quality generations on CIFAR-10 and LSUN bedrooms.

An evaluation of direct attacks using fake fingers generated from ISO templates This work reports a vulnerability evaluation of a highly competitive ISO matcher to direct attacks carried out with fake fingers generated from ISO templates. Experiments are carried out on a fingerprint database acquired in a real-life scenario and show that the evaluated system is highly vulnerable to the proposed attack scheme, granting access in over 75% of the attempts (for a high-security operating point). Thus, the study disproves the popular belief of minutiae templates non-reversibility and raises a key vulnerability issue in the use of non-encrypted standard templates. (This article is an extended version of Galbally et al., 2008, which was awarded with the IBM Best Student Paper Award in the track of Biometrics at ICPR 2008).

Safe mutations for deep and recurrent neural networks through output gradients While neuroevolution (evolving neural networks) has been successful across a variety of domains from reinforcement learning, to artificial life, to evolutionary robotics, it is rarely applied to large, deep neural networks. A central reason is that while random mutation generally works in low dimensions, a random perturbation of thousands or millions of weights will likely break existing functionality. This paper proposes a solution: a family of safe mutation (SM) operators that facilitate exploration without dramatically altering network behavior or requiring additional interaction with the environment. The most effective SM variant scales the degree of mutation of each individual weight according to the sensitivity of the network's outputs to that weight, which requires computing the gradient of outputs with respect to the weights (instead of the gradient of error, as in conventional deep learning). This safe mutation through gradients (SM-G) operator dramatically increases the ability of a simple genetic algorithm-based neuroevolution method to find solutions in high-dimensional domains that require deep and/or recurrent neural networks, including domains that require processing raw pixels. By improving our ability to evolve deep neural networks, this new safer approach to mutation expands the scope of domains amenable to neuroevolution.

Synthesizing the preferred inputs for neurons in neural networks via deep generator networks. Deep neural networks (DNNs) have demonstrated state-of-the-art results on many pattern recognition tasks, especially vision classification problems. Understanding the inner workings of such computational brains is both fascinating basic science that is interesting in its own right-similar to why we study the human brain-and will enable researchers to further improve DNNs. One path to understanding how a neural network functions internally is to study what each of its neurons has learned to detect. One such method is called activation maximization (AM), which synthesizes an input (e.g. an image) that highly activates a neuron. Here we dramatically improve the qualitative state of the art of activation maximization by harnessing a powerful, learned prior: a deep generator network (DGN). The algorithm (1) generates qualitatively state-of-the-art synthetic images that look almost real, (2) reveals the features learned by each neuron in an interpretable way, (3) generalizes well to new datasets and somewhat well to different network architectures without requiring the prior to be relearned, and (4) can be considered as a high-quality generative method (in this case, by generating novel, creative, interesting, recognizable images).

Learning To Learn Relation For Important People Detection In Still Images Humans can easily recognize the importance of people in social event images, and they always focus on the most important individuals. However, learning to learn the relation between people in an image, and inferring the most important person based on this relation, remains undeveloped. In this work, we propose a deep imPOrtance relatIon NeTwork (POINT) that combines both relation modeling and feature learning. In particular, we infer two types of interaction modules: the person-person interaction module that learns the interaction between people and the event-person interaction module that learns to describe how a person is involved in the event occurring in an image. We then estimate the importance relations among people from both interactions and encode the relation feature from the importance relations. In this way, POINT automatically learns several types of relation features in parallel, and we aggregate these relation features and the person's feature to form the importance feature for important people classification. Extensive experimental results show that our method is effective for important people detection and verify the efficacy of learning to learn relations for important people detection.

Unsupervised Pixel-Level Domain Adaptation with Generative Adversarial Networks Collecting well-annotated image datasets to train modern machine learning algorithms is prohibitively expensive for many tasks. One appealing alternative is rendering synthetic data where ground-truth annotations are generated automatically. Unfortunately, models trained purely on rendered images fail to generalize to real images. To address this shortcoming, prior work introduced unsupervised domain adaptation algorithms that have tried to either map representations between the two domains, or learn to extract features that are domain-invariant. In this work, we approach the problem in a new light by learning in an unsupervised manner a transformation in the pixel space from one domain to the other. Our generative adversarial network (GAN)-based method adapts source-domain images to appear as if drawn from the target domain. Our approach not only produces plausible samples, but also outperforms the state-of-the-art on a number of unsupervised domain adaptation scenarios by large margins. Finally, we demonstrate that the adaptation process generalizes to object classes unseen during training.

Switching between stabilizing controllers This paper deals with the problem of switching between several linear time-invariant (LTI) controllers—all of them capable of stabilizing a speci4c LTI process—in such a way that the stability of the closed-loop system is guaranteed for any switching sequence. We show that it is possible to 4nd realizations for any given family of controller transfer matrices so that the closed-loop system remains stable, no matter how we switch among the controller. The motivation for this problem is the control of complex systems where con8icting requirements make a single LTI controller unsuitable. ? 2002 Published by Elsevier Science Ltd.

A General Equilibrium Model for Industries with Price and Service Competition This paper develops a stochastic general equilibrium inventory model for an oligopoly, in which all inventory constraint parameters are endogenously determined. We propose several systems of demand processes whose distributions are functions of all retailers' prices and all retailers' service levels. We proceed with the investigation of the equilibrium behavior of infinite-horizon models for industries facing this type of generalized competition, under demand uncertainty.We systematically consider the following three competition scenarios. (1) Price competition only: Here, we assume that the firms' service levels are exogenously chosen, but characterize how the price and inventory strategy equilibrium vary with the chosen service levels. (2) Simultaneous price and service-level competition: Here, each of the firms simultaneously chooses a service level and a combined price and inventory strategy. (3) Two-stage competition: The firms make their competitive choices sequentially. In a first stage, all firms simultaneously choose a service level; in a second stage, the firms simultaneously choose a combined pricing and inventory strategy with full knowledge of the service levels selected by all competitors. We show that in all of the above settings a Nash equilibrium of infinite-horizon stationary strategies exists and that it is of a simple structure, provided a Nash equilibrium exists in a so-called reduced game.We pay particular attention to the question of whether a firm can choose its service level on the basis of its own (input) characteristics (i.e., its cost parameters and demand function) only. We also investigate under which of the demand models a firm, under simultaneous competition, responds to a change in the exogenously specified characteristics of the various competitors by either: (i) adjusting its service level and price in the same direction, thereby compensating for price increases (decreases) by offering improved (inferior) service, or (ii) adjusting them in opposite directions, thereby simultaneously offering better or worse prices and service.

Improved Schemes for Visual Cryptography A (k,n)-threshold visual cryptography scheme ((k,n)-threshold VCS, for short) is a method to encode a secret image SI into n shadow images called shares such that any k or more shares enable the “visual” recovery of the secret image, but by inspecting less than k shares one cannot gain any information on the secret image. The “visual” recovery consists of xeroxing the shares onto transparencies, and then stacking them. Any k shares will reveal the secret image without any cryptographic computation.In this paper we analyze visual cryptography schemes in which the reconstruction of black pixels is perfect, that is, all the subpixels associated to a black pixel are black. For any value of k and n, where 2\leq k\leq n, we give a construction for (k,n)-threshold VCS which improves on the best previously known constructions with respect to the pixel expansion (i.e., the number of subpixels each pixel of the original image is encoded into). We also provide a construction for coloured (2,n)-threshold VCS and for coloured (n,n)-threshold VCS. Both constructions improve on the best previously known constructions with respect to the pixel expansion.

FreeTagpaper: a pen-and-paper-based collaboration system using visual tags printed on paper We propose a novel collaboration system using pen and paper. FreeTagpaper uses a camera set above a work table to recognize the visual tags printed on the paper to acquire the paper's ID and posture. From the paper ID, FreeTagpaper can retrieve the paper's size and format from a document management database. It then uses the format information to transmit images of handwritten notes to the other side and simultaneously projects the other side's handwritten notes onto this paper. Based on this mutual image sharing scheme, users can collaborate in real time using paper of any size, format, and posture. This report describes preliminary work on FreeTagpaper including the current implementation and essential experimental results.

Distributed Kalman consensus filter with event-triggered communication: Formulation and stability analysis. •The problem of distributed state estimation in sensor networks with event-triggered communication schedules on both sensor-to-estimator channel and estimator-to-estimator channel is studied.•An event-triggered KCF is designed by deriving the optimal Kalman gain matrix which minimizes the mean squared error.•A computational scalable form of the proposed filter is presented by some approximations.•An appropriate choice of the consensus gain matrix is provided to ensure the stochastic stability of the proposed filter.

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

SPDS: A Secure and Auditable Private Data Sharing Scheme for Smart Grid Based on Blockchain The exponential growth of data generated from increasing smart meters and smart appliances brings about huge potentials for more efficient energy production, pricing, and personalized energy services in smart grids. However, it also causes severe concerns due to improper use of individuals' private data, as well as the lack of transparency and auditability for data usage. To bridge this gap, in this article, we propose a secure and auditable private data sharing (SPDS) scheme under data processing-as-a-service mode in smart grid. Specifically, we first present a novel blockchain-based framework for trust-free private data computation and data usage tracking, where smart contracts are employed to specify fine-grained data usage policies (i.e., who can access what kinds of data, for what purposes, at what price) while the distributed ledgers keep an immutable and transparent record of data usage. A trusted execution environment based off-chain smart contract execution mechanism is exploited as well to process confidential user datasets and relieve the computation overhead in blockchain systems. A two-phase atomic delivery protocol is designed to ensure the atomicity of data transactions in computing result release and payment. Furthermore, based on contract theory, the optimal contracts are designed under information asymmetry to stimulate user's participation and high-quality data sharing while optimizing the payoff of the energy service provider. Extensive simulation results demonstrate that the proposed SPDS can effectively improve the payoffs of participants, compared with conventional schemes.

Privacy Enabled Digital Rights Management Without Trusted Third Party Assumption Digital rights management systems are required to provide security and accountability without violating the privacy of the entities involved. However, achieving privacy along with accountability in the same framework is hard as these attributes are mutually contradictory. Thus, most of the current digital rights management systems rely on trusted third parties to provide privacy to the entities involved. However, a trusted third party can become malicious and break the privacy protection of the entities in the system. Hence, in this paper, we propose a novel privacy preserving content distribution mechanism for digital rights management without relying on the trusted third party assumption. We use simple primitives such as blind decryption and one way hash chain to avoid the trusted third party assumption. We prove that our scheme is not prone to the “oracle problem” of the blind decryption mechanism. The proposed mechanism supports access control without degrading user's privacy as well as allows revocation of even malicious users without violating their privacy.

An efficient conditionally anonymous ring signature in the random oracle model A conditionally anonymous ring signature is an exception since the anonymity is conditional. Specifically, it allows an entity to confirm/refute the signature that he generated before. A group signature also shares the same property since a group manager can revoke a signer's anonymity using the trapdoor information. However, the special node (i.e., group manager) does not exist in the group in order to satisfy the ad hoc fashion. In this paper, we construct a new conditionally anonymous ring signature, in which the actual signer can be traced without the help of the group manager. The big advantage of the confirmation and disavowal protocols designed by us are non-interactive with constant costs while the known schemes suffer from the linear cost in terms of the ring size n or security parameter s.

Threats to Networking Cloud and Edge Datacenters in the Internet of Things. Several application domains are collecting data using Internet of Things sensing devices and shipping it to remote cloud datacenters for analysis (fusion, storage, and processing). Data analytics activities raise a new set of technical challenges from the perspective of ensuring end-to-end security and privacy of data as it travels from an edge datacenter (EDC) to a cloud datacenter (CDC) (or vice...

Performance Analysis of the Raft Consensus Algorithm for Private Blockchains Consensus is one of the key problems in blockchains. There are many articles analyzing the performance of threat models for blockchains. But the network stability seems lack of attention, which in fact affects the blockchain performance. This paper studies the performance of a well adopted consensus algorithm, Raft, in networks with non-negligible packet loss rate. In particular, we propose a simple but accurate analytical model to analyze the distributed network split probability. At a given time, we explicitly present the network split probability as a function of the network size, the packet loss rate, and the election timeout period. To validate our analysis, we implement a Raft simulator and the simulation results coincide with the analytical results. With the proposed model, one can predict the network split time and probability in theory and optimize the parameters in Raft consensus algorithm.

A Personalized Privacy Protection Framework for Mobile Crowdsensing in IIoT With the rapid digitalization of various industries, mobile crowdsensing (MCS), an intelligent data collection and processing paradigm of the industrial Internet of Things, has provided a promising opportunity to construct powerful industrial systems and provide industrial services. The existing unified privacy strategy for all sensing data results in excessive or insufficient protection and low quality of crowdsensing services (QoCS) in MCS. To tackle this issue, in this article we propose a personalized privacy protection (PERIO) framework based on game theory and data encryption. Initially, we design a personalized privacy measurement algorithm to calculate users’ privacy level, which is then combined with game theory to construct a rational uploading strategy. Furthermore, we propose a privacy-preserving data aggregation scheme to ensure data confidentiality, integrity, and real-timeness. Theoretical analysis and ample simulations with real trajectory dataset indicate that the PERIO scheme is effective and makes a reasonable balance between retaining high QoCS and privacy.

Chaos-Based Content Distribution Framework for Digital Rights Management System Multimedia contents are digitally utilized these days. Thus, the development of an effective method to access the content is becoming the topmost priority of the entertainment industry to protect the digital content from unauthorized access. Digital rights management (DRM) systems are the technique that makes digital content accessible only to the legal rights holders. As the Internet of Things environment is used in the distribution and access of digital content, a secure and efficient content delivery mechanism is also required. Keeping the focus on these points, this article proposes a content distribution framework for DRM system using chaotic map. Formal security verification under the random oracle model, which uncovers the proposed protocol's capability to resist the critical attacks is given. Moreover, simulation study for security verification is performed using the broadly accepted “automated validation of Internet security protocols and applications,” which indicates that the protocol is safe. Moreover, the detailed comparative study with related protocols demonstrates that it provides better security and improves the computational and communication efficiency.

Constrained Kalman filtering for indoor localization of transport vehicles using floor-installed HF RFID transponders Localization of transport vehicles is an important issue for many intralogistics applications. The paper presents an inexpensive solution for indoor localization of vehicles. Global localization is realized by detection of RFID transponders, which are integrated in the floor. The paper presents a novel algorithm for fusing RFID readings with odometry using Constraint Kalman filtering. The paper presents experimental results with a Mecanum based omnidirectional vehicle on a NaviFloor® installation, which includes passive HF RFID transponders. The experiments show that the proposed Constraint Kalman filter provides a similar localization accuracy compared to a Particle filter but with much lower computational expense.

Constrained Multiobjective Optimization for IoT-Enabled Computation Offloading in Collaborative Edge and Cloud Computing Internet-of-Things (IoT) applications are becoming more resource-hungry and latency-sensitive, which are severely constrained by limited resources of current mobile hardware. Mobile cloud computing (MCC) can provide abundant computation resources, while mobile-edge computing (MEC) aims to reduce the transmission latency by offloading complex tasks from IoT devices to nearby edge servers. It is sti...

MOEA/D: A Multiobjective Evolutionary Algorithm Based on Decomposition Decomposition is a basic strategy in traditional multiobjective optimization. However, it has not yet been widely used in multiobjective evolutionary optimization. This paper proposes a multiobjective evolutionary algorithm based on decomposition (MOEA/D). It decomposes a multiobjective optimization problem into a number of scalar optimization subproblems and optimizes them simultaneously. Each subproblem is optimized by only using information from its several neighboring subproblems, which makes MOEA/D have lower computational complexity at each generation than MOGLS and nondominated sorting genetic algorithm II (NSGA-II). Experimental results have demonstrated that MOEA/D with simple decomposition methods outperforms or performs similarly to MOGLS and NSGA-II on multiobjective 0-1 knapsack problems and continuous multiobjective optimization problems. It has been shown that MOEA/D using objective normalization can deal with disparately-scaled objectives, and MOEA/D with an advanced decomposition method can generate a set of very evenly distributed solutions for 3-objective test instances. The ability of MOEA/D with small population, the scalability and sensitivity of MOEA/D have also been experimentally investigated in this paper.

A Model for Understanding How Virtual Reality Aids Complex Conceptual Learning Designers and evaluators of immersive virtual reality systems have many ideas concerning how virtual reality can facilitate learning. However, we have little information concerning which of virtual reality's features provide the most leverage for enhancing understanding or how to customize those affordances for different learning environments. In part, this reflects the truly complex nature of learning. Features of a learning environment do not act in isolation; other factors such as the concepts or skills to be learned, individual characteristics, the learning experience, and the interaction experience all play a role in shaping the learning process and its outcomes. Through Project Science Space, we have been trying to identify, use, and evaluate immersive virtual reality's affordances as a means to facilitate the mastery of complex, abstract concepts. In doing so, we are beginning to understand the interplay between virtual reality's features and other important factors in shaping the learning process and learning outcomes for this type of material. In this paper, we present a general model that describes how we think these factors work together and discuss some of the lessons we are learning about virtual reality's affordances in the context of this model for complex conceptual learning.

Solving the data sparsity problem in destination prediction Destination prediction is an essential task for many emerging location-based applications such as recommending sightseeing places and targeted advertising according to destinations. A common approach to destination prediction is to derive the probability of a location being the destination based on historical trajectories. However, almost all the existing techniques use various kinds of extra information such as road network, proprietary travel planner, statistics requested from government, and personal driving habits. Such extra information, in most circumstances, is unavailable or very costly to obtain. Thereby we approach the task of destination prediction by using only historical trajectory dataset. However, this approach encounters the \"data sparsity problem\", i.e., the available historical trajectories are far from enough to cover all possible query trajectories, which considerably limits the number of query trajectories that can obtain predicted destinations. We propose a novel method named Sub-Trajectory Synthesis (SubSyn) to address the data sparsity problem. SubSyn first decomposes historical trajectories into sub-trajectories comprising two adjacent locations, and then connects the sub-trajectories into \"synthesised\" trajectories. This process effectively expands the historical trajectory dataset to contain much more trajectories. Experiments based on real datasets show that SubSyn can predict destinations for up to ten times more query trajectories than a baseline prediction algorithm. Furthermore, the running time of the SubSyn-training algorithm is almost negligible for a large set of 1.9 million trajectories, and the SubSyn-prediction algorithm runs over two orders of magnitude faster than the baseline prediction algorithm constantly.

Design of robust fuzzy fault detection filter for polynomial fuzzy systems with new finite frequency specifications This paper investigates the problem of fault detection filter design for discrete-time polynomial fuzzy systems with faults and unknown disturbances. The frequency ranges of the faults and the disturbances are assumed to be known beforehand and to reside in low, middle or high frequency ranges. Thus, the proposed filter is designed in the finite frequency range to overcome the conservatism generated by those designed in the full frequency domain. Being of polynomial fuzzy structure, the proposed filter combines the H−/H∞ performances in order to ensure the best robustness to the disturbance and the best sensitivity to the fault. Design conditions are derived in Sum Of Squares formulations that can be easily solved via available software tools. Two illustrative examples are introduced to demonstrate the effectiveness of the proposed method and a comparative study with LMI method is also provided.

Intention-detection strategies for upper limb exosuits: model-based myoelectric vs dynamic-based control The cognitive human-robot interaction between an exosuit and its wearer plays a key role in determining both the biomechanical effects of the device on movements and its perceived effectiveness. There is a lack of evidence, however, on the comparative performance of different control methods, implemented on the same device. Here, we compare two different control approaches on the same robotic suit: a model-based myoelectric control (myoprocessor), which estimates the joint torque from the activation of target muscles, and a dynamic-based control that provides support against gravity using an inverse dynamic model. Tested on a cohort of four healthy participants, assistance from the exosuit results in a marked reduction in the effort of muscles working against gravity with both control approaches (peak reduction of 68.6±18.8%, for the dynamic arm model and 62.4±25.1% for the myoprocessor), when compared to an unpowered condition. Neither of the two controllers had an affect on the performance of their users in a joint-angle tracking task (peak errors of 15.4° and 16.4° for the dynamic arm model and myoprocessor, respectively, compared to 13.1o in the unpowered condition). However, our results highlight the remarkable adaptability of the myoprocessor to seamlessly adapt to changing external dynamics.

Learning Discriminative Features with Multiple Granularities for Person Re-Identification. The combination of global and partial features has been an essential solution to improve discriminative performances in person re-identification (Re-ID) tasks. Previous part-based methods mainly focus on locating regions with specific pre-defined semantics to learn local representations, which increases learning difficulty but not efficient or robust to scenarios with large variances. In this paper, we propose an end-to-end feature learning strategy integrating discriminative information with various granularities. We carefully design the Multiple Granularity Network (MGN), a multi-branch deep network architecture consisting of one branch for global feature representations and two branches for local feature representations. Instead of learning on semantic regions, we uniformly partition the images into several stripes, and vary the number of parts in different local branches to obtain local feature representations with multiple granularities. Comprehensive experiments implemented on the mainstream evaluation datasets including Market-1501, DukeMTMC-reid and CUHK03 indicate that our method robustly achieves state-of-the-art performances and outperforms any existing approaches by a large margin. For example, on Market-1501 dataset in single query mode, we obtain a top result of Rank-1/mAP=96.6%/94.2% with this method after re-ranking.

High-Order Information Matters: Learning Relation And Topology For Occluded Person Re-Identification Occluded person re-identification (ReID) aims to match occluded person images to holistic ones across disjoint cameras. In this paper, we propose a novel framework by learning high-order relation and topology information for discriminative features and robust alignment. At first, we use a CNN backbone and a key-points estimation model to extract semantic local features. Even so, occluded images still suffer from occlusion and outliers. Then, we view the local features of an image as nodes of a graph and propose an adaptive direction graph convolutional (ADGC) layer to pass relation information between nodes. The proposed ADGC layer can automatically suppress the message passing of meaningless features by dynamically learning direction and degree of linkage. When aligning two groups of local features from two images, we view it as a graph matching problem and propose a cross-graph embedded-alignment (CGEA) layer to jointly learn and embed topology information to local features, and straightly predict similarity score. The proposed CGEA layer not only take full use of alignment learned by graph matching but also replace sensitive one-to-one matching with a robust soft one. Finally, extensive experiments on occluded, partial, and holistic ReID tasks show the effectiveness of our proposed method. Specifically, our framework significantly outperforms state-of-the-art by 6.5% mAP scores on Occluded-Duke dataset. Code is available at https://github.com/wangguanan/HOReID.

Perceive Where to Focus: Learning Visibility-Aware Part-Level Features for Partial Person Re-Identification This paper considers a realistic problem in person re-identification (re-ID) task, i.e., partial re-ID. Under partial re-ID scenario, the images may contain a partial observation of a pedestrian. If we directly compare a partial pedestrian image with a holistic one, the extreme spatial misalignment significantly compromises the discriminative ability of the learned representation. We propose a Visibility-aware Part Model (VPM) for partial re-ID, which learns to perceive the visibility of regions through self-supervision. The visibility awareness allows VPM to extract region-level features and compare two images with focus on their shared regions (which are visible on both images). VPM gains two-fold benefit toward higher accuracy for partial re-ID. On the one hand, compared with learning a global feature, VPM learns region-level features and thus benefits from fine-grained information. On the other hand, with visibility awareness, VPM is capable to estimate the shared regions between two images and thus suppresses the spatial misalignment. Experimental results confirm that our method significantly improves the learned feature representation and the achieved accuracy is on par with the state of the art.

Vrstc: Occlusion-Free Video Person Re-Identification Video person re-identification (re-ID) plays an important role in surveillance video analysis. However, the performance of video re-ID degenerates severely under partial occlusion. In this paper, we propose a novel network, called Spatio-Temporal Completion network (STCnet), to explicitly handle partial occlusion problem. Different from most previous works that discard the occluded frames, STCnet can recover the appearance of the occluded parts. For one thing, the spatial structure of a pedestrian frame can be used to predict the occluded body parts from the unoccluded body parts of this frame. For another, the temporal patterns of pedestrian sequence provide important clues to generate the contents of occluded parts. With the spatio-temporal information, STCnet can recover the appearance for the occluded parts, which could be leveraged with those unoccluded parts for more accurate video re-ID. By combining a re-ID network with STCnet, a video re-ID framework robust to partial occlusion (VRSTC) is proposed. Experiments on three challenging video re-ID databases demonstrate that the proposed approach outperforms the state-of-the-arts.

Group Reidentification with Multigrained Matching and Integration The task of reidentifying groups of people under different camera views is an important yet less-studied problem. Group reidentification (Re-ID) is a very challenging task since it is not only adversely affected by common issues in traditional single-object Re-ID problems, such as viewpoint and human pose variations, but also suffers from changes in group layout and group membership. In this paper, we propose a novel concept of group granularity by characterizing a group image by multigrained objects: individual people and subgroups of two and three people within a group. To achieve robust group Re-ID, we first introduce multigrained representations which can be extracted via the development of two separate schemes, that is, one with handcrafted descriptors and another with deep neural networks. The proposed representation seeks to characterize both appearance and spatial relations of multigrained objects, and is further equipped with importance weights which capture variations in intragroup dynamics. Optimal group-wise matching is facilitated by a multiorder matching process which, in turn, dynamically updates the importance weights in iterative fashion. We evaluated three multicamera group datasets containing complex scenarios and large dynamics, with experimental results demonstrating the effectiveness of our approach.

Pose-Guided Feature Alignment for Occluded Person Re-Identification Persons are often occluded by various obstacles in person retrieval scenarios. Previous person re-identification (re-id) methods, either overlook this issue or resolve it based on an extreme assumption. To alleviate the occlusion problem, we propose to detect the occluded regions, and explicitly exclude those regions during feature generation and matching. In this paper, we introduce a novel method named Pose-Guided Feature Alignment (PGFA), exploiting pose landmarks to disentangle the useful information from the occlusion noise. During the feature constructing stage, our method utilizes human landmarks to generate attention maps. The generated attention maps indicate if a specific body part is occluded and guide our model to attend to the non-occluded regions. During matching, we explicitly partition the global feature into parts and use the pose landmarks to indicate which partial features belonging to the target person. Only the visible regions are utilized for the retrieval. Besides, we construct a large-scale dataset for the Occluded Person Re-ID problem, namely Occluded-DukeMTMC, which is by far the largest dataset for the Occlusion Person Re-ID. Extensive experiments are conducted on our constructed occluded re-id dataset, two partial re-id datasets, and two commonly used holistic re-id datasets. Our method largely outperforms existing person re-id methods on three occlusion datasets, while remains top performance on two holistic datasets.

Deep Learning for Person Re-Identification: A Survey and Outlook Person re-identification (Re-ID) aims at retrieving a person of interest across multiple non-overlapping cameras. With the advancement of deep neural networks and increasing demand of intelligent video surveillance, it has gained significantly increased interest in the computer vision community. By dissecting the involved components in developing a person Re-ID system, we categorize it into the cl...

A Survey on Transfer Learning A major assumption in many machine learning and data mining algorithms is that the training and future data must be in the same feature space and have the same distribution. However, in many real-world applications, this assumption may not hold. For example, we sometimes have a classification task in one domain of interest, but we only have sufficient training data in another domain of interest, where the latter data may be in a different feature space or follow a different data distribution. In such cases, knowledge transfer, if done successfully, would greatly improve the performance of learning by avoiding much expensive data-labeling efforts. In recent years, transfer learning has emerged as a new learning framework to address this problem. This survey focuses on categorizing and reviewing the current progress on transfer learning for classification, regression, and clustering problems. In this survey, we discuss the relationship between transfer learning and other related machine learning techniques such as domain adaptation, multitask learning and sample selection bias, as well as covariate shift. We also explore some potential future issues in transfer learning research.

A Low-Complexity Analytical Modeling for Cross-Layer Adaptive Error Protection in Video Over WLAN We find a low-complicity and accurate model to solve the problem of optimizing MAC-layer transmission of real-time video over wireless local area networks (WLANs) using cross-layer techniques. The objective in this problem is to obtain the optimal MAC retry limit in order to minimize the total packet loss rate. First, the accuracy of Fluid and M/M/1/K analytical models is examined. Then we derive a closed-form expression for service time in WLAN MAC transmission, and will use this in mathematical formulation of our optimization problem based on M/G/1 model. Subsequently we introduce an approximate and simple formula for MAC-layer service time, which leads to the M/M/1 model. Compared with M/G/1, we particularly show that our M/M/1-based model provides a low-complexity and yet quite accurate means for analyzing MAC transmission process in WLAN. Using our M/M/1 model-based analysis, we derive closed-form formulas for the packet overflow drop rate and optimum retry-limit. These closed-form expressions can be effectively invoked for analyzing adaptive retry-limit algorithms. Simulation results (network simulator-2) will verify the accuracy of our analytical models.

Toward a theory of intrinsically motivating instruction First, a number of previous theories of intrinsic motivation are reviewed. Then, several studies of highly motivating computer games are described. These studies focus on what makes the games fun, not on what makes them educational. Finally, with this background, a rudimentary theory of intrinsically motivating instruction is developed, based on three categories: challenge, fantasy, and curiosity. Challenge is hypothesized to depend on goals with uncertain outcomes. Several ways of making outcomes uncertain are discussed, including variable difficulty level, multiple level goals, hidden information, and randomness. Fantasy is claimed to have both cognitive and emotional advantages in designing instructional environments. A distinction is made between extrinsic fantasies that depend only weakly on the skill used in a game, and intrinsic fantasies that are intimately related to the use of the skill. Curiosity is separated into sensory and cognitive components, and it is suggested that cognitive curiosity can be aroused by making learners believe their knowledge structures are incomplete, inconsistent, or unparsimonious.

Constrained Interaction and Coordination in Proximity-Limited Multiagent Systems In this paper, we consider the problem of controlling the interactions of a group of mobile agents, subject to a set of topological constraints. Assuming proximity-limited interagent communication, we leverage mobility, unlike prior work, to enable adjacent agents to interact discriminatively, i.e., to actively retain or reject communication links on the basis of constraint satisfaction. Specifically, we propose a distributed scheme that consists of hybrid controllers with discrete switching for link discrimination, coupled with attractive and repulsive potentials fields for mobility control, where constraint violation predicates form the basis for discernment. We analyze the application of constrained interaction to two canonical coordination objectives, i.e., aggregation and dispersion, with maximum and minimum node degree constraints, respectively. For each task, we propose predicates and control potentials, and examine the dynamical properties of the resulting hybrid systems. Simulation results demonstrate the correctness of our proposed methods and the ability of our framework to generate topology-aware coordinated behavior.

Distributed adaptive containment control of uncertain nonlinear multi-agent systems in strict-feedback form. This paper presents a distributed containment control approach for uncertain nonlinear strict-feedback systems with multiple dynamic leaders under a directed graph topology where the leaders are neighbors of only a subset of the followers. The strict-feedback followers with nonparametric uncertainties are considered and the local adaptive dynamic surface controller for each follower is designed using only neighbors’ information to guarantee that all followers converge to the dynamic convex hull spanned by the dynamic leaders where the derivatives of leader signals are not available to implement controllers, i.e., the position information of leaders is only required. The function approximation technique using neural networks is employed to estimate nonlinear uncertainty terms derived from the controller design procedure for the followers. It is shown that the containment control errors converge to an adjustable neighborhood of the origin.

Privacy-preserving Naive Bayes classifiers secure against the substitution-then-comparison attack. Naive Bayes (NB) is a simple but highly practical classifier, with a wide range of applications including spam filters, cancer diagnosis and face recognition, to name a few examples only. Consider a situation where a user requests a classification service from a NB classifier server, both the user and the server do not want to reveal their private data to each other. This paper focuses on constructing a privacy-preserving NB classifier that is resistant to an easy-to-perform, but difficult-to-detect attack, which we call the substitution-then-comparison (STC) attack. Without resorting to fully homomorphic encryptions, which has a high computational overhead, we propose a scheme which avoids information leakage under the STC attack. Our key technique involves the use of a “double-blinding” technique, and we show how to combine it with additively homomorphic encryptions and oblivious transfer to hide both parties’ privacy. Furthermore, a completed evaluation shows that the construction is highly practical - most of the computations are in the server’s offline phase, and the overhead of online computation and communication is small for both parties.

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

A new approach to solving the multiple traveling salesperson problem using genetic algorithms The multiple traveling salesperson problem (MTSP) involves scheduling m>1 salespersons to visit a set of n>m locations so that each location is visited exactly once while minimizing the total (or maximum) distance traveled by the salespersons. The MTSP is similar to the notoriously difficult traveling salesperson problem (TSP) with the added complication that each location may be visited by any one of the salespersons. Previous studies investigated solving the MTSP with genetic algorithms (GAs) using standard TSP chromosomes and operators. This paper proposes a new GA chromosome and related operators for the MTSP and compares the theoretical properties and computational performance of the proposed technique to previous work. Computational testing shows the new approach results in a smaller search space and, in many cases, produces better solutions than previous techniques.

Touring a sequence of polygons Given a sequence of k polygons in the plane, a start point s, and a target point, t, we seek a shortest path that starts at s, visits in order each of the polygons, and ends at t. If the polygons are disjoint and convex, we give an algorithm running in time O(kn log (n/k)), where n is the total number of vertices specifying the polygons. We also extend our results to a case in which the convex polygons are arbitrarily intersecting and the subpath between any two consecutive polygons is constrained to lie within a simply connected region; the algorithm uses O(nk2 log n) time. Our methods are simple and allow shortest path queries from s to a query point t to be answered in time O(k log n + m), where m is the combinatorial path length. We show that for nonconvex polygons this "touring polygons" problem is NP-hard.The touring polygons problem is a strict generalization of some classic problems in computational geometry, including the safari problem, the zoo-keeper problem, and the watchman route problem in a simple polygon. Our new results give an order of magnitude improvement in the running times of the safari problem and the watchman route problem: We solve the safari problem in O(n2 log n) time and the watchman route problem (through a fixed point s) in time O(n3 log n), compared with the previous time bounds of O(n3) and O(n4), respectively.

Numerical Comparison of Some Penalty-Based Constraint Handling Techniques in Genetic Algorithms We study five penalty function-based constraint handling techniques to be used with genetic algorithms in global optimization. Three of them, the method of superiority of feasible points, the method of parameter free penalties and the method of adaptive penalties have already been considered in the literature. In addition, we introduce two new modifications of these methods. We compare all the five methods numerically in 33 test problems and report and analyze the results obtained in terms of accuracy, efficiency and reliability. The method of adaptive penalties turned out to be most efficient while the method of parameter free penalties was the most reliable.

Well-Solvable Special Cases of the Traveling Salesman Problem: A Survey. The traveling salesman problem (TSP) belongs to the most basic, most important, and most investigated problems in combinatorial optimization. Although it is an ${\cal NP}$-hard problem, many of its special cases can be solved efficiently in polynomial time. We survey these special cases with emphasis on the results that have been obtained during the decade 1985--1995. This survey complements an earlier survey from 1985 compiled by Gilmore, Lawler, and Shmoys [The Traveling Salesman Problem---A Guided Tour of Combinatorial Optimization, Wiley, Chichester, pp. 87--143].

Rich Vehicle Routing Problem: Survey The Vehicle Routing Problem (VRP) is a well-known research line in the optimization research community. Its different basic variants have been widely explored in the literature. Even though it has been studied for years, the research around it is still very active. The new tendency is mainly focused on applying this study case to real-life problems. Due to this trend, the Rich VRP arises: combining multiple constraints for tackling realistic problems. Nowadays, some studies have considered specific combinations of real-life constraints to define the emerging Rich VRP scopes. This work surveys the state of the art in the field, summarizing problem combinations, constraints defined, and approaches found.

Implementing Vehicle Routing Algorithms

Communication in reactive multiagent robotic systems Multiple cooperating robots are able to complete many tasks more quickly and reliably than one robot alone. Communication between the robots can multiply their capabilities and effectiveness, but to what extent? In this research, the importance of communication in robotic societies is investigated through experiments on both simulated and real robots. Performance was measured for three different types of communication for three different tasks. The levels of communication are progressively more complex and potentially more expensive to implement. For some tasks, communication can significantly improve performance, but for others inter-agent communication is apparently unnecessary. In cases where communication helps, the lowest level of communication is almost as effective as the more complex type. The bulk of these results are derived from thousands of simulations run with randomly generated initial conditions. The simulation results help determine appropriate parameters for the reactive control system which was ported for tests on Denning mobile robots.

Crowd sensing of traffic anomalies based on human mobility and social media The advances in mobile computing and social networking services enable people to probe the dynamics of a city. In this paper, we address the problem of detecting and describing traffic anomalies using crowd sensing with two forms of data, human mobility and social media. Traffic anomalies are caused by accidents, control, protests, sport events, celebrations, disasters and other events. Unlike existing traffic-anomaly-detection methods, we identify anomalies according to drivers' routing behavior on an urban road network. Here, a detected anomaly is represented by a sub-graph of a road network where drivers' routing behaviors significantly differ from their original patterns. We then try to describe the detected anomaly by mining representative terms from the social media that people posted when the anomaly happened. The system for detecting such traffic anomalies can benefit both drivers and transportation authorities, e.g., by notifying drivers approaching an anomaly and suggesting alternative routes, as well as supporting traffic jam diagnosis and dispersal. We evaluate our system with a GPS trajectory dataset generated by over 30,000 taxicabs over a period of 3 months in Beijing, and a dataset of tweets collected from WeiBo, a Twitter-like social site in China. The results demonstrate the effectiveness and efficiency of our system.

Joint Optimization of Radio and Computational Resources for Multicell Mobile-Edge Computing Migrating computational intensive tasks from mobile devices to more resourceful cloud servers is a promising technique to increase the computational capacity of mobile devices while saving their battery energy. In this paper, we consider a MIMO multicell system where multiple mobile users (MUs) ask for computation offloading to a common cloud server. We formulate the offloading problem as the joint optimization of the radio resources􀀀the transmit precoding matrices of the MUs􀀀and the computational resources􀀀the CPU cycles/second assigned by the cloud to each MU􀀀in order to minimize the overall users’ energy consumption, while meeting latency constraints. The resulting optimization problem is nonconvex (in the objective function and constraints). Nevertheless, in the single-user case, we are able to compute the global optimal solution in closed form. In the more challenging multiuser scenario, we propose an iterative algorithm, based on a novel successive convex approximation technique, converging to a local optimal solution of the original nonconvex problem. We then show that the proposed algorithmic framework naturally leads to a distributed and parallel implementation across the radio access points, requiring only a limited coordination/signaling with the cloud. Numerical results show that the proposed schemes outperform disjoint optimization algorithms.

A bayesian network approach to traffic flow forecasting A new approach based on Bayesian networks for traffic flow forecasting is proposed. In this paper, traffic flows among adjacent road links in a transportation network are modeled as a Bayesian network. The joint probability distribution between the cause nodes (data utilized for forecasting) and the effect node (data to be forecasted) in a constructed Bayesian network is described as a Gaussian mixture model (GMM) whose parameters are estimated via the competitive expectation maximization (CEM) algorithm. Finally, traffic flow forecasting is performed under the criterion of minimum mean square error (mmse). The approach departs from many existing traffic flow forecasting models in that it explicitly includes information from adjacent road links to analyze the trends of the current link statistically. Furthermore, it also encompasses the issue of traffic flow forecasting when incomplete data exist. Comprehensive experiments on urban vehicular traffic flow data of Beijing and comparisons with several other methods show that the Bayesian network is a very promising and effective approach for traffic flow modeling and forecasting, both for complete data and incomplete data

Cognitive Cars: A New Frontier for ADAS Research This paper provides a survey of recent works on cognitive cars with a focus on driver-oriented intelligent vehicle motion control. The main objective here is to clarify the goals and guidelines for future development in the area of advanced driver-assistance systems (ADASs). Two major research directions are investigated and discussed in detail: 1) stimuli–decisions–actions, which focuses on the driver side, and 2) perception enhancement–action-suggestion–function-delegation, which emphasizes the ADAS side. This paper addresses the important achievements and major difficulties of each direction and discusses how to combine the two directions into a single integrated system to obtain safety and comfort while driving. Other related topics, including driver training and infrastructure design, are also studied.

Biologically-inspired soft exosuit. In this paper, we present the design and evaluation of a novel soft cable-driven exosuit that can apply forces to the body to assist walking. Unlike traditional exoskeletons which contain rigid framing elements, the soft exosuit is worn like clothing, yet can generate moments at the ankle and hip with magnitudes of 18% and 30% of those naturally generated by the body during walking, respectively. Our design uses geared motors to pull on Bowden cables connected to the suit near the ankle. The suit has the advantages over a traditional exoskeleton in that the wearer's joints are unconstrained by external rigid structures, and the worn part of the suit is extremely light, which minimizes the suit's unintentional interference with the body's natural biomechanics. However, a soft suit presents challenges related to actuation force transfer and control, since the body is compliant and cannot support large pressures comfortably. We discuss the design of the suit and actuation system, including principles by which soft suits can transfer force to the body effectively and the biological inspiration for the design. For a soft exosuit, an important design parameter is the combined effective stiffness of the suit and its interface to the wearer. We characterize the exosuit's effective stiffness, and present preliminary results from it generating assistive torques to a subject during walking. We envision such an exosuit having broad applicability for assisting healthy individuals as well as those with muscle weakness.

Inter-class sparsity based discriminative least square regression Least square regression is a very popular supervised classification method. However, two main issues greatly limit its performance. The first one is that it only focuses on fitting the input features to the corresponding output labels while ignoring the correlations among samples. The second one is that the used label matrix, i.e., zero–one label matrix is inappropriate for classification. To solve these problems and improve the performance, this paper presents a novel method, i.e., inter-class sparsity based discriminative least square regression (ICS_DLSR), for multi-class classification. Different from other methods, the proposed method pursues that the transformed samples have a common sparsity structure in each class. For this goal, an inter-class sparsity constraint is introduced to the least square regression model such that the margins of samples from the same class can be greatly reduced while those of samples from different classes can be enlarged. In addition, an error term with row-sparsity constraint is introduced to relax the strict zero–one label matrix, which allows the method to be more flexible in learning the discriminative transformation matrix. These factors encourage the method to learn a more compact and discriminative transformation for regression and thus has the potential to perform better than other methods. Extensive experimental results show that the proposed method achieves the best performance in comparison with other methods for multi-class classification.

Social Robots for (Second) Language Learning in (Migrant) Primary School Children Especially these days, innovation and support from technology to relieve pressure in education is highly urgent. This study tested the potential advantage of a social robot over a tablet in (second) language learning on performance, engagement, and enjoyment. Shortages in primary education call for new technology solutions. Previous studies combined robots with tablets, to compensate for robot’s limitations, however, this study applied direct human–robot interaction. Primary school children (N = 63, aged 4–6) participated in a 3-wave field experiment with story-telling exercises, either with a semi-autonomous robot (without tablet, using WOz) or a tablet. Results showed increased learning gains over time when training with a social robot, compared to the tablet. Children who trained with a robot were more engaged in the story-telling task and enjoyed it more. Robot’s behavioral style (social or neutral) hardly differed overall, however, seems to vary for high versus low educational abilities. While social robots need sophistication before being implemented in schools, our study shows the potential of social robots as tutors in (second) language learning.

On the distance constrained vehicle routing problem We analyze the vehicle routing problem with constraints on the total distance traveled by each vehicle. Two objective functions are considered: minimize the total distance traveled by vehicles and ...

Approximation algorithms for some vehicle routing problems We study vehicle routing problems with constraints on the distance traveled by each vehicle or on the number of vehicles. The objective is either to minimize the total distance traveled by vehicles or to minimize the number of vehicles used. We design constant differential approximation algorithms for kVRP. Note that, using the differential bound for METRIC 3VRP, we obtain the randomized standard ratio 197/99 + ε ∀ε 0. This is an improvement of the best-known bound of 2 given by Haimovich et al. (Vehicle Routing Methods and Studies, Golden, Assad, editors, Elsevier, Amsterdam, 1988). For natural generalizations of this problem, called EDGE COST VRP, VERTEX COST VRP, MIN VEHICLE and kTSP we obtain constant differential approximation algorithms and we show that these problems have no differential approximation scheme, unless P = NP.

Touring a sequence of polygons Given a sequence of k polygons in the plane, a start point s, and a target point, t, we seek a shortest path that starts at s, visits in order each of the polygons, and ends at t. If the polygons are disjoint and convex, we give an algorithm running in time O(kn log (n/k)), where n is the total number of vertices specifying the polygons. We also extend our results to a case in which the convex polygons are arbitrarily intersecting and the subpath between any two consecutive polygons is constrained to lie within a simply connected region; the algorithm uses O(nk2 log n) time. Our methods are simple and allow shortest path queries from s to a query point t to be answered in time O(k log n + m), where m is the combinatorial path length. We show that for nonconvex polygons this "touring polygons" problem is NP-hard.The touring polygons problem is a strict generalization of some classic problems in computational geometry, including the safari problem, the zoo-keeper problem, and the watchman route problem in a simple polygon. Our new results give an order of magnitude improvement in the running times of the safari problem and the watchman route problem: We solve the safari problem in O(n2 log n) time and the watchman route problem (through a fixed point s) in time O(n3 log n), compared with the previous time bounds of O(n3) and O(n4), respectively.

New exact method for large asymmetric distance-constrained vehicle routing problem. In this paper we revise and modify an old branch-and-bound method for solving the asymmetric distance-constrained vehicle routing problem suggested by Laporte et al. in 1987. Our modification is based on reformulating distance-constrained vehicle routing problem into a travelling salesman problem, and on using assignment problem as a lower bounding procedure. In addition, our algorithm uses the best-first strategy and new tolerance based branching rules. Since our method is fast but memory consuming, it could stop before optimality is proven. Therefore, we introduce the randomness, in case of ties, in choosing the node of the search tree. If an optimal solution is not found, we restart our procedure. As far as we know, the instances that we have solved exactly (up to 1000 customers) are much larger than the instances considered for other vehicle routing problem models from the recent literature. So, despite of its simplicity, this proposed algorithm is capable of solving the largest instances ever solved in the literature. Moreover, this approach is general and may be used for solving other types of vehicle routing problems. (C) 2012 Elsevier B.V. All rights reserved.

A Survey on Vehicle Routing Problem with Loading Constraints In the classical vehicle routing problem (VRP), a fleet of vehicles is available to serve a set of customers with known demand. Each customer is visited by exactly one vehicle as well as the objective is to minimize the total distance or the total charge incurred. Recent years have seen increased attention on VRP integrated with additional loading constraints, known as 2L-CVRP or 3L-CVRP. In this paper we present a survey of the state-of-the-art on 2L-CVRP/3L-CVRP.

Implementing Vehicle Routing Algorithms

Wireless Charging Technologies: Fundamentals, Standards, and Network Applications Wireless charging is a technology of transmitting power through an air gap to electrical devices for the purpose of energy replenishment. The recent progress in wireless charging techniques and development of commercial products have provided a promising alternative way to address the energy bottleneck of conventionally portable battery-powered devices. However, the incorporation of wireless charg...

Network Under Limited Mobile Devices: A New Technique for Mobile Charging Scheduling With Multiple Sinks. Recently, many studies have investigated scheduling mobile devices to recharge and collect data from sensors in wireless rechargeable sensor networks (WRSNs) such that the network lifetime is prolonged. In reality, because mobile devices are more powerful and expensive than sensors, the cost of the mobile devices often consumes a high portion of the budget. Due to a limited budget, the number of m...

Avoiding radiation of on-demand multi-node energy charging with multiple mobile chargers. Wireless rechargeable sensor networks (WRSN) exploit wireless energy transfer techniques to replenish sensor batteries using chargers. Compared to non-rechargeable networks, WRSNs offer controllable and predictable energy replenishment and prolong the network lifetime. While previous mobile charging protocols in WRSNs consider multi-node energy transfer or the use of several chargers, in this work, we propose combining the two approaches. Our solution leverages simultaneous energy transfer to multiple nodes, with the aim of maximizing the number of nodes that are charged at each stop to reduce the charging delay. Moreover, it considers multiple chargers to serve more requesting nodes, which is expected to limit the number of node failures. Furthermore, we guarantee that there is no aggregated electromagnetic radiation (EMR), which can occur when two nearby chargers are simultaneously charging. This has not yet been considered in a mobile environment. To address these challenges, we introduce a new clustering algorithm to group demanding nodes, and we compute charging tours by iteratively solving a Traveling Salesman Problem with Multiple Time Windows problem (TSPMTW). The simulation results demonstrate that our approach reduces node failures in dynamic networks.

Neural Architecture Transfer Neural architecture search (NAS) has emerged as a promising avenue for automatically designing task-specific neural networks. Existing NAS approaches require one complete search for each deployment specification of hardware or objective. This is a computationally impractical endeavor given the potentially large number of application scenarios. In this paper, we propose Neural Architecture ...

VSNR: A Wavelet-Based Visual Signal-to-Noise Ratio for Natural Images This paper presents an efficient metric for quantifying the visual fidelity of natural images based on near-threshold and suprathreshold properties of human vision. The proposed metric, the visual signal-to-noise ratio (VSNR), operates via a two-stage approach. In the first stage, contrast thresholds for detection of distortions in the presence of natural images are computed via wavelet-based models of visual masking and visual summation in order to determine whether the distortions in the distorted image are visible. If the distortions are below the threshold of detection, the distorted image is deemed to be of perfect visual fidelity (VSNR = infin)and no further analysis is required. If the distortions are suprathreshold, a second stage is applied which operates based on the low-level visual property of perceived contrast, and the mid-level visual property of global precedence. These two properties are modeled as Euclidean distances in distortion-contrast space of a multiscale wavelet decomposition, and VSNR is computed based on a simple linear sum of these distances. The proposed VSNR metric is generally competitive with current metrics of visual fidelity; it is efficient both in terms of its low computational complexity and in terms of its low memory requirements; and it operates based on physical luminances and visual angle (rather than on digital pixel values and pixel-based dimensions) to accommodate different viewing conditions.

Joint Optimization of Source Precoding and Relay Beamforming in Wireless MIMO Relay Networks. This paper considers joint linear processing at multi-antenna sources and one multiple-input multiple-output (MIMO) relay station for both one-way and two-way relay-assisted wireless communications. The one-way relaying is applicable in the scenario of downlink transmission by a multi-antenna base station to multiple single-antenna users with the help of one MIMO relay. In such a scenario, the objective of join linear processing is to maximize the information throughput to users. The design problem is equivalently formulated as the maximization of the worst signal-to-interference-plus-noise ratio (SINR) among all users subject to various transmission power constraints. Such a program of nonconvex objective minimization under nonconvex constraints is transformed to a canonical d.c. (difference of convex functions/sets) program of d.c. function optimization under convex constraints through nonconvex duality with zero duality gap. An efficient iterative algorithm is then applied to solve this canonical d.c program. For the scenario of using one MIMO relay to assist two sources exchanging their information in two-way relying manner, the joint linear processing aims at either minimizing the maximum mean square error (MSE) or maximizing the total information throughput of the two sources. By applying tractable optimization for the linear minimum MSE estimator and d.c. programming, an iterative algorithm is developed to solve these two optimization problems. Extensive simulation results demonstrate that the proposed methods substantially outperform previously-known joint optimization methods.

Massive MIMO Antenna Selection: Switching Architectures, Capacity Bounds, and Optimal Antenna Selection Algorithms. Antenna selection is a multiple-input multiple-output (MIMO) technology, which uses radio frequency (RF) switches to select a good subset of antennas. Antenna selection can alleviate the requirement on the number of RF transceivers, thus being attractive for massive MIMO systems. In massive MIMO antenna selection systems, RF switching architectures need to be carefully considered. In this paper, w...

STNReID: Deep Convolutional Networks With Pairwise Spatial Transformer Networks for Partial Person Re-Identification Partial person re-identification (ReID) is a challenging task because only partial information of person images is available for matching target persons. Few studies, especially on deep learning, have focused on matching partial person images with holistic person images. This study presents a novel deep partial ReID framework based on pairwise spatial transformer networks (STNReID), which can be trained on existing holistic person datasets. STNReID includes a spatial transformer network (STN) module and a ReID module. The STN module samples an affined image (a semantically corresponding patch) from the holistic image to match the partial image. The ReID module extracts the features of the holistic, partial, and affined images. Competition (or confrontation) is observed between the STN module and the ReID module, and two-stage training is applied to acquire a strong STNReID for partial ReID. Experimental results show that our STNReID obtains 66.7% and 54.6% rank-1 accuracies on Partial-ReID and Partial-iLIDS datasets, respectively. These values are at par with those obtained with state-of-the-art methods.

DeepFace: Closing the Gap to Human-Level Performance in Face Verification In modern face recognition, the conventional pipeline consists of four stages: detect => align => represent => classify. We revisit both the alignment step and the representation step by employing explicit 3D face modeling in order to apply a piecewise affine transformation, and derive a face representation from a nine-layer deep neural network. This deep network involves more than 120 million parameters using several locally connected layers without weight sharing, rather than the standard convolutional layers. Thus we trained it on the largest facial dataset to-date, an identity labeled dataset of four million facial images belonging to more than 4, 000 identities. The learned representations coupling the accurate model-based alignment with the large facial database generalize remarkably well to faces in unconstrained environments, even with a simple classifier. Our method reaches an accuracy of 97.35% on the Labeled Faces in the Wild (LFW) dataset, reducing the error of the current state of the art by more than 27%, closely approaching human-level performance.

A survey on deep learning based face recognition. Deep learning, in particular the deep convolutional neural networks, has received increasing interests in face recognition recently, and a number of deep learning methods have been proposed. This paper summarizes about 330 contributions in this area. It reviews major deep learning concepts pertinent to face image analysis and face recognition, and provides a concise overview of studies on specific face recognition problems, such as handling variations in pose, age, illumination, expression, and heterogeneous face matching. A summary of databases used for deep face recognition is given as well. Finally, some open challenges and directions are discussed for future research.

Generating Master Faces for Use in Performing Wolf Attacks on Face Recognition Systems Due to its convenience, biometric authentication, especial face authentication, has become increasingly mainstream and thus is now a prime target for attackers. Presentation attacks and face morphing are typical types of attack. Previous research has shown that finger- vein- and fingerprint-based authentication methods are susceptible to wolf attacks, in which a wolf sample matches many enrolled user templates. In this work, we demonstrated that wolf (generic) faces, which we call “master faces,” can also compromise face recognition systems and that the master face concept can be generalized in some cases. Motivated by recent similar work in the fingerprint domain, we generated high-quality master faces by using the state-of-the-art face generator StyleGAN in a process called latent variable evolution. Experiments demonstrated that even attackers with limited resources using only pre-trained models available on the Internet can initiate master face attacks. The results, in addition to demonstrating performance from the attacker's point of view, can also be used to clarify and improve the performance of face recognition systems and harden face authentication systems.

The Effect of Broad and Specific Demographic Homogeneity on the Imposter Distributions and False Match Rates in Face Recognition Algorithm Performance The growing adoption of biometric identity systems, notably face recognition, has raised questions regarding whether performance is equitable across demographic groups. Prior work on this issue showed that performance of face recognition systems varies with demographic variables. However, biometric systems make two distinct types of matching errors, which lead to different outcomes for users depending on the technology use case. In this research, we develop a framework for classifying biometric performance differentials that separately considers the effect of false positive and false negative outcomes, and show that oft-cited evidence regarding biometric equitability has focused on primarily on false-negatives. We then correlate demographic variables with false-positive outcomes in a diverse population using a commercial face recognition algorithm, and show that false match rate (FMR) at a fixed threshold increases >400-fold for broadly homogeneous groups (individuals of the same age, same gender, and same race) relative to heterogeneous groups. This was driven by systematic shifts in the tails of the imposter distribution impelled primarily by homogeneity in race and gender. For specific demographic groups, we observed the highest false match rate for older males that self-identified as White and the lowest for older males that self-identified as Black or African American. The magnitude of FMR differentials between specific homogeneous groups (<;3-fold) was modest in comparison with the FMR increase associated with broad demographic homogeneity. These results demonstrate the false positive outcomes of face recognition systems are not simply linked to single demographic factors, and that a careful consideration of interactions between multiple factors is needed when considering the equitability of these systems.

The FERET Evaluation Methodology for Face-Recognition Algorithms Two of the most critical requirements in support of producing reliable face-recognition systems are a large database of facial images and a testing procedure to evaluate systems. The Face Recognition Technology (FERET) program has addressed both issues through the FERET database of facial images and the establishment of the FERET tests. To date, 14,126 images from 1,199 individuals are included in the FERET database, which is divided into development and sequestered portions of the database. In September 1996, the FERET program administered the third in a series of FERET face-recognition tests. The primary objectives of the third test were to 1) assess the state of the art, 2) identify future areas of research, and 3) measure algorithm performance.

Auto-Alert: A Spatial and Temporal Architecture for Driving Assistance in Road Traffic Environments Over the last decade, the Advanced Driver Assistance System (ADAS) concept has evolved prominently. ADAS involves several advanced approaches such as automotive electronics, vehicular communication, RADAR, LIDAR, computer vision, and its associated aspects such as machine learning and deep learning. Of these, computer vision and machine learning-based solutions have mainly been effective that have allowed real-time vehicle control, driver-aided systems, etc. However, most of the existing works deal with ADAS deployment and autonomous driving functionality in countries with well-disciplined lane traffic. These solutions and frameworks do not work in countries and cities with less-disciplined/ chaotic traffic. Hence, critical ADAS functionalities and even L2/ L3 autonomy levels in driving remain a major open challenge. In this regard, this work proposes a novel framework called Auto-Alert. Auto-Alert performs a two-stage spatial and temporal analysis based on external traffic environment and tri-axial sensor system for safe driving assistance. This work investigates time-series analysis with deep learning models for driving events prediction and assistance. Further, as a basic premise, various essential design considerations towards the ADAS are discussed. Significantly, the Convolutional Neural Network (CNN) and Long-Short-Term-Memory (LSTM) models are applied in the proposed Auto-Alert. It is shown that the LSTM outperforms the CNN with 99% for the considered window length. Importantly, this also involves developing and demonstrating an efficient traffic monitoring and density estimation system. Further, this work provides the benchmark results for Indian Driving Dataset (IDD), specifically for the object detection task. The findings of this proposed work demonstrate the significance of using CNN and LSTM networks to assist the driver in the holistic traffic environment.

Modeling, Identification, and Predictive Control of a Driver Steering Assistance System. This paper presents the design of a driver steering assistance system, which provides a corrective torque in order to guide the driver. While designing such a system, it is important to consider the interactions since the driver modifies the transfer function from the control input to the output of interest during a shared steering task. The novelty of our approach lies in the formulation of the p...

Trust in Automation: Integrating Empirical Evidence on Factors That Influence Trust. Objective: We systematically review recent empirical research on factors that influence trust in automation to present a three-layered trust model that synthesizes existing knowledge. Background: Much of the existing research on factors that guide human-automation interaction is centered around trust, a variable that often determines the willingness of human operators to rely on automation. Studies have utilized a variety of different automated systems in diverse experimental paradigms to identify factors that impact operators' trust. Method: We performed a systematic review of empirical research on trust in automation from January 2002 to June 2013. Papers were deemed eligible only if they reported the results of a human-subjects experiment in which humans interacted with an automated system in order to achieve a goal. Additionally, a relationship between trust (or a trust-related behavior) and another variable had to be measured. All together, 101 total papers, containing 127 eligible studies, were included in the review. Results: Our analysis revealed three layers of variability in human-automation trust (dispositional trust, situational trust, and learned trust), which we organize into a model. We propose design recommendations for creating trustworthy automation and identify environmental conditions that can affect the strength of the relationship between trust and reliance. Future research directions are also discussed for each layer of trust. Conclusion: Our three-layered trust model provides a new lens for conceptualizing the variability of trust in automation. Its structure can be applied to help guide future research and develop training interventions and design procedures that encourage appropriate trust.

Real-Time Driver Maneuver Prediction Using LSTM Driver maneuver prediction is of great importance in designing a modern Advanced Driver Assistance System (ADAS). Such predictions can improve driving safety by alerting the driver to the danger of unsafe or risky traffic situations. In this research, we developed a model to predict driver maneuvers, including left/right lane changes, left/right turns and driving straight forward 3.6 seconds on average before they occur in real time. For this, we propose a deep learning method based on Long Short-Term Memory (LSTM) which utilizes data on the driver's gaze and head position as well as vehicle dynamics data. We applied our approach on real data collected during drives in an urban environment in an instrumented vehicle. In comparison with previous IOHMM techniques that predicted three maneuvers including left/right turns and driving straight, our prediction model is able to anticipate two more maneuvers. In addition to this, our experimental results show that our model using identical dataset improved F1 score by 4% and increased to 84%.

Dual Attention Network For Scene Segmentation In this paper, we address the scene segmentation task by capturing rich contextual dependencies based on the self-attention mechanism. Unlike previous works that capture contexts by multi-scale feature fusion, we propose a Dual Attention Network (DANet) to adaptively integrate local features with their global dependencies. Specifically, we append two types of attention modules on top of dilated FCN, which model the semantic interdependencies in spatial and channel dimensions respectively. The position attention module selectively aggregates the feature at each position by a weighted sum of the features at all positions. Similar features would be related to each other regardless of their distances. Meanwhile, the channel attention module selectively emphasizes interdependent channel maps by integrating associated features among all channel maps. We sum the outputs of the two attention modules to further improve feature representation which contributes to more precise segmentation results. We achieve new state-of-the-art segmentation performance on three challenging scene segmentation datasets, i.e., Cityscapes, PASCAL Context and COCO Stuff dataset. In particular, a Mean IoU score of 81.5% on Cityscapes test set is achieved without using coarse data.(1).

Secure High-Rate Transaction Processing in Bitcoin. Bitcoin is a disruptive new crypto-currency based on a decentralized open-source protocol which has been gradually gaining momentum. Perhaps the most important question that will affect Bitcoin's success, is whether or not it will be able to scale to support the high volume of transactions required from a global currency system. We investigate the implications of having a higher transaction throughput on Bitcoin's security against double-spend attacks. We show that at high throughput, substantially weaker attackers are able to reverse payments they have made, even well after they were considered accepted by recipients. We address this security concern through the GHOST rule, a modification to the way Bitcoin nodes construct and re-organize the block chain, Bitcoin's core distributed data-structure. GHOST has been adopted and a variant of it has been implemented as part of the Ethereum project, a second generation distributed applications platform.

Design, Implementation, and Experimental Results of a Quaternion-Based Kalman Filter for Human Body Motion Tracking Real-time tracking of human body motion is an important technology in synthetic environments, robotics, and other human-computer interaction applications. This paper presents an extended Kalman filter designed for real-time estimation of the orientation of human limb segments. The filter processes data from small inertial/magnetic sensor modules containing triaxial angular rate sensors, accelerometers, and magnetometers. The filter represents rotation using quaternions rather than Euler angles or axis/angle pairs. Preprocessing of the acceleration and magnetometer measurements using the Quest algorithm produces a computed quaternion input for the filter. This preprocessing reduces the dimension of the state vector and makes the measurement equations linear. Real-time implementation and testing results of the quaternion-based Kalman filter are presented. Experimental results validate the filter design, and show the feasibility of using inertial/magnetic sensor modules for real-time human body motion tracking

Neural network-based event-triggered MFAC for nonlinear discrete-time processes. This paper is concerned with the event-triggered data-driven control problem for nonlinear discrete-time systems. An event-based data-driven model-free adaptive controller design algorithm together with constructing an adaptive event-trigger condition is developed. Different from the existing data-driven model-free adaptive control approach, an aperiodic neural network weight update law is introduced to estimate the controller parameters, and the event-trigger mechanism is activated only if the event-trigger error exceeds the threshold. Furthermore, by combining the equivalent-dynamic-linearization technique with the Lyapunov method, it is proved that both the closed-loop control system and the weight estimation error are ultimately bounded. Finally, two simulation examples are provided to demonstrate the effectiveness of the derived method.

Communication-Efficient Federated Learning Over MIMO Multiple Access Channels Communication efficiency is of importance for wireless federated learning systems. In this paper, we propose a communication-efficient strategy for federated learning over multiple-input multiple-output (MIMO) multiple access channels (MACs). The proposed strategy comprises two components. When sending a locally computed gradient, each device compresses a high dimensional local gradient to multiple lower-dimensional gradient vectors using block sparsification. When receiving a superposition of the compressed local gradients via a MIMO-MAC, a parameter server (PS) performs a joint MIMO detection and the sparse local-gradient recovery. Inspired by the turbo decoding principle, our joint detection-and-recovery algorithm accurately recovers the high-dimensional local gradients by iteratively exchanging their beliefs for MIMO detection and sparse local gradient recovery outputs. We then analyze the reconstruction error of the proposed algorithm and its impact on the convergence rate of federated learning. From simulations, our gradient compression and joint detection-and-recovery methods diminish the communication cost significantly while achieving identical classification accuracy for the case without any compression.

A memetic algorithm with competition for the capacitated green vehicle routing problem In this paper, a memetic algorithm with competition ( MAC ) is proposed to solve the capacitated green vehicle routing problem ( CGVRP ). Firstly, the permutation array called traveling salesman problem ( TSP ) route is used to encode the solution, and an effective decoding method to construct the CGVRP route is presented accordingly. Secondly, the k-nearest neighbor ( kNN ) based initialization is presented to take use of the location information of the customers. Thirdly, according to the characteristics of the CGVRP, the search operators in the variable neighborhood search ( VNS ) framework and the simulated annealing ( SA ) strategy are executed on the TSP route for all solutions. Moreover, the customer adjustment operator and the alternative fuel station ( AFS ) adjustment operator on the CGVRP route are executed for the elite solutions after competition. In addition, the crossover operator is employed to share information among different solutions. The effect of parameter setting is investigated using the Taguchi method of design-of-experiment to suggest suitable values. Via numerical tests, it demonstrates the effectiveness of both the competitive search and the decoding method. Moreover, extensive comparative results show that the proposed algorithm is more effective and efficient than the existing methods in solving the CGVRP.

People detection and tracking from aerial thermal views Detection and tracking of people in visible-light images has been subject to extensive research in the past decades with applications ranging from surveillance to search-and-rescue. Following the growing availability of thermal cameras and the distinctive thermal signature of humans, research effort has been focusing on developing people detection and tracking methodologies applicable to this sensing modality. However, a plethora of challenges arise on the transition from visible-light to thermal images, especially with the recent trend of employing thermal cameras onboard aerial platforms (e.g. in search-and-rescue research) capturing oblique views of the scenery. This paper presents a new, publicly available dataset of annotated thermal image sequences, posing a multitude of challenges for people detection and tracking. Moreover, we propose a new particle filter based framework for tracking people in aerial thermal images. Finally, we evaluate the performance of this pipeline on our dataset, incorporating a selection of relevant, state-of-the-art methods and present a comprehensive discussion of the merits spawning from our study.

Pareto-Optimization for Scheduling of Crude Oil Operations in Refinery via Genetic Algorithm. With the interaction of discrete-event and continuous processes, it is challenging to schedule crude oil operations in a refinery. This paper studies the optimization problem of finding a detailed schedule to realize a given refining schedule. This is a multiobjective optimization problem with a combinatorial nature. Since the original problem cannot be directly solved by using heuristics and meta-heuristics, the problem is transformed into an assignment problem of charging tanks and distillers. Based on such a transformation, by analyzing the properties of the problem, this paper develops a chromosome that can describe a feasible schedule such that meta-heuristics can be applied. Then, it innovatively adopts an improved nondominated sorting genetic algorithm to solve the problem for the first time. An industrial case study is used to test the proposed solution method. The results show that the method makes a significant performance improvement and is applicable to real-life refinery scheduling problems.

The unmanned aerial vehicle routing and trajectory optimisation problem, a taxonomic review. •We introduce the UAV Routing and Trajectory Optimisation Problem.•We provide a taxonomy for UAV routing, TO and other variants.•We apply the proposed taxonomy to 70 scientific articles.•A lack of research about integrating UAV routing and TO is identified.

A vehicle routing problem arising in unmanned aerial monitoring. •We consider a routing problem arising in unmanned aerial monitoring.•The problem possesses several military and civil applications.•The problem is to construct vehicle routes and to select monitoring heights.•We model the problem as an integer linear program and solve it by tabu search.•Extensive tests confirm the efficiency of the heuristic.

Analysis of Mixed-Integer Linear Programming Formulations for a Fuel-Constrained Multiple Vehicle Routing Problem This paper addresses a fuel-constrained, multiple vehicle routing problem (FCMVRP) in the presence of multiple refueling stations. We are given a set of targets, a set of refueling stations, and a depot where m vehicles are stationed. The vehicles are allowed to refuel at any refueling station, and the objective of the problem is to determine a route for each vehicle starting and terminating at the depot, such that each target is visited by at least one vehicle, the vehicles never run out of fuel while traversing their routes, and the total travel cost of all the routes is a minimum. We present four new mixed-integer linear programming (MILP) formulations for the problem. These formulations are compared both analytically and empirically, and a branch-and-cut algorithm is developed to compute an optimal solution. Extensive computational results on a large class of test instances that corroborate the effectiveness of the algorithm are also presented.

A fitness approximation assisted competitive swarm optimizer for large scale expensive optimization problems. Surrogate assisted meta-heuristic algorithms have received increasing attention over the past years due to the fact that many real-world optimization problems are computationally expensive. However, most existing surrogate assisted meta-heuristic algorithms are designed for small or medium scale problems. In this paper, a fitness approximation assisted competitive swarm optimizer is proposed for optimization of large scale expensive problems. Different from most surrogate assisted evolutionary algorithms that use a computational model for approximating the fitness, we estimate the fitness based on the positional relationship between individuals in the competitive swarm optimizer. Empirical study on seven widely used benchmark problems with 100 and 500 decision variables show that the proposed fitness approximation assisted competitive swarm optimizer is able to achieve competitive performance on a limited computational budget.

Autoencoding Evolutionary Search With Learning Across Heterogeneous Problems. To enhance the search performance of evolutionary algorithms, reusing knowledge captured from past optimization experiences along the search process has been proposed in the literature, and demonstrated much promise. In the literature, there are generally three types of approaches for reusing knowledge from past search experiences, namely exact storage and reuse of past solutions, the reuse of mod...

Probabilistic encryption A new probabilistic model of data encryption is introduced. For this model, under suitable complexity assumptions, it is proved that extracting any information about the cleartext from the cyphertext is hard on the average for an adversary with polynomially bounded computational resources. The proof holds for any message space with any probability distribution. The first implementation of this model is presented. The security of this implementation is proved under the interactability assumptin of deciding Quadratic Residuosity modulo composite numbers whose factorization is unknown.

Argos: practical many-antenna base stations Multi-user multiple-input multiple-output theory predicts manyfold capacity gains by leveraging many antennas on wireless base stations to serve multiple clients simultaneously through multi-user beamforming (MUBF). However, realizing a base station with a large number antennas is non-trivial, and has yet to be achieved in the real-world. We present the design, realization, and evaluation of Argos, the first reported base station architecture that is capable of serving many terminals simultaneously through MUBF with a large number of antennas (M >> 10). Designed for extreme flexibility and scalability, Argos exploits hierarchical and modular design principles, properly partitions baseband processing, and holistically considers real-time requirements of MUBF. Argos employs a novel, completely distributed, beamforming technique, as well as an internal calibration procedure to enable implicit beamforming with channel estimation cost independent of the number of base station antennas. We report an Argos prototype with 64 antennas and capable of serving 15 clients simultaneously. We experimentally demonstrate that by scaling from 1 to 64 antennas the prototype can achieve up to 6.7 fold capacity gains while using a mere 1/64th of the transmission power.

Redundancy, Efficiency and Robustness in Multi-Robot Coverage ó Area coverage is an important task for mobile robots, with many real-world applications. Motivated by poten- tial efciency and robustness improvements, there is growing interest in the use of multiple robots in coverage. Previous investigations of multi-robot coverage focuses on completeness and eliminating redundancy, but does not formally address ro- bustness, nor examine the impact of the initial positions of robots on the coverage time. Indeed, a common assumption is that non-redundancy leads to improved coverage time. We address robustness and efciency in a family of multi-robot coverage algorithms, based on spanning-tree coverage of approximate cell decomposition. We analytically show that the algorithms are robust, in that as long as a single robot is able to move, the coverage will be completed. We also show that non-redundant (non-backtracking) versions of the algorithms have a worst-case coverage time virtually identical to that of a single robotó thus no performance gain is guaranteed in non-redundant coverage. Moreover, this worst-case is in fact common in real- world applications. Surprisingly, however, redundant coverage algorithms lead to guaranteed performance which halves the coverage time even in the worst case. produces a path that completely covers the work-area. We want multi-robot algorithms to be not only complete, but also efcient (in that they minimize the time it takes to cover the area), non-backtracking (in that any portion of the work area is covered only once), and robust (in that they can handle catastrophic robot failures). Previous investigations that examine the use of multiple robots in coverage mostly focus on completeness and non- backtracking. However, much of previous work does not formally consider robustness. Moreover, while completeness and non-backtracking properties are sufcient to show that a single-robot coverage algorithm is also efcient (in cov- erage time), it turns out that this is not true in the general case. Surprisingly, in multi-robot coverage, non-backtracking and efcienc y are independent optimization criteria: Non- backtracking algorithms may be inefcient, and efcient algorithms may be backtracking. Finally, the initial position of robots in the work-area signicantly affects the comple- tion time of the coverage, both in backtracking and non- backtracking algorithms. Yet no bounds are known for the coverage completion time, as a function of the number of robots and their initial placement. This paper examines robustness and efcienc y in multi- robot coverage. We focus on coverage using a map of the work-area (known as off-line coverage (1)). We assume the tool to be a square of size D. The work-area is then approximately decomposed into cells, where each cell is a square of size 4D, i.e., a square of four tool-size sub-cells. As with other approximate cell-decomposition approaches ((1)), cells that are partially coveredóby obstacles or the bounds of the work-areaóare discarded from consideration. We use an algorithm based on a spanning-tree to extract a path that visits all sub-cells. Previous work on generating such a path (called STC for Spanning-Tree Coverage) have shown it to be complete and non-backtracking (3). We present a family of novel algorithms, called MSTC (Multirobot Spanning-Tree Coverage) that address robustness and efcienc y. First, we construct a non-backtracking MSTC algorithm that is guaranteed to be robust: It guarantees that the work-area will be completely covered in nite time, as long as at least a single robot is functioning correctly. We

Heterogeneous ensemble for feature drifts in data streams The nature of data streams requires classification algorithms to be real-time, efficient, and able to cope with high-dimensional data that are continuously arriving. It is a known fact that in high-dimensional datasets, not all features are critical for training a classifier. To improve the performance of data stream classification, we propose an algorithm called HEFT-Stream (H eterogeneous E nsemble with F eature drifT for Data Streams ) that incorporates feature selection into a heterogeneous ensemble to adapt to different types of concept drifts. As an example of the proposed framework, we first modify the FCBF [13] algorithm so that it dynamically update the relevant feature subsets for data streams. Next, a heterogeneous ensemble is constructed based on different online classifiers, including Online Naive Bayes and CVFDT [5]. Empirical results show that our ensemble classifier outperforms state-of-the-art ensemble classifiers (AWE [15] and OnlineBagging [21]) in terms of accuracy, speed, and scalability. The success of HEFT-Stream opens new research directions in understanding the relationship between feature selection techniques and ensemble learning to achieve better classification performance.

An ID-Based Linearly Homomorphic Signature Scheme and Its Application in Blockchain. Identity-based cryptosystems mean that public keys can be directly derived from user identifiers, such as telephone numbers, email addresses, and social insurance number, and so on. So they can simplify key management procedures of certificate-based public key infrastructures and can be used to realize authentication in blockchain. Linearly homomorphic signature schemes allow to perform linear computations on authenticated data. And the correctness of the computation can be publicly verified. Although a series of homomorphic signature schemes have been designed recently, there are few homomorphic signature schemes designed in identity-based cryptography. In this paper, we construct a new ID-based linear homomorphic signature scheme, which avoids the shortcomings of the use of public-key certificates. The scheme is proved secure against existential forgery on adaptively chosen message and ID attack under the random oracle model. The ID-based linearly homomorphic signature schemes can be applied in e-business and cloud computing. Finally, we show how to apply it to realize authentication in blockchain.

Hardware Circuits Design and Performance Evaluation of a Soft Lower Limb Exoskeleton Soft lower limb exoskeletons (LLEs) are wearable devices that have good potential in walking rehabilitation and augmentation. While a few studies focused on the structure design and assistance force optimization of the soft LLEs, rarely work has been conducted on the hardware circuits design. The main purpose of this work is to present a new soft LLE for walking efficiency improvement and introduce its hardware circuits design. A soft LLE for hip flexion assistance and a hardware circuits system with scalability were proposed. To assess the efficacy of the soft LLE, the experimental tests that evaluate the sensor data acquisition, force tracking performance, lower limb muscle activity and metabolic cost were conducted. The time error in the peak assistance force was just 1%. The reduction in the normalized root-mean-square EMG of the rectus femoris was 7.1%. The net metabolic cost in exoskeleton on condition was reduced by 7.8% relative to walking with no exoskeleton. The results show that the designed hardware circuits can be applied to the soft LLE and the soft LLE is able to improve walking efficiency of wearers.

Anthropomorphic Movement Analysis and Synthesis: A Survey of Methods and Applications. The anthropomorphic body form is a complex articulated system of links/limbs and joints, simultaneously redundant and underactuated, and capable of a wide range of sophisticated movement. The human body and its movement have long been a topic of study in physiology, anatomy, biomechanics, and neuroscience and have served as inspiration for humanoid robot design and control. This survey paper revie...

Parametrization and Range of Motion of the Ball-and-Socket Joint The ball-and-socket joint model is used to represent articulations with three rotational degrees of free- dom (DOF), such as the human shoulder and the hip. The goal of this paper is to discuss two related prob- lems: the parametrization and the definition of realistic joint boundaries for ball-and-socket joints. Doing this accurately is difficult, yet important for motion generators (such as inverse kinematics and dynamics engines) and for motion manipulators (such as motion retargeting), since the resulting motions should satisfy the anatomic constraints. The difficulty mainly comes from the complex nature of 3D orientations and of human articulations. The underlying question of parametrization must be addressed before realis- tic and meaningful boundaries can be defined over the set of 3D orientations. In this paper, we review and compare several known methods, and advocate the use of the swing-and-twist parametrization, that parti- tions an arbitrary orientation into two meaningful components. The related problem of induced twist is discussed. Finally, we review some joint boundaries representations based on this decomposition, and show an example.

Positional kinematics of humanoid arms We present the positional abilities of a humanoid manipulator based on an improved kinematical model of the human arm. This was synthesized from electro-optical measurements of healthy female and male subjects. The model possesses three joints: inner shoulder joint, outer shoulder joint and elbow joint. The first functions as the human sternoclavicular joint, the second functions as the human glenohumeral joint, and the last replicates the human humeroulnar rotation. There are three links included, the forearm and the upper arm link which are of a constant length, and the shoulder link which is expandable. Mathematical interrelations between the joint coordinates are also taken into consideration. We determined the reachability of a humanoid arm, treated its orienting redundancy in the shoulder complex and the positional redundancy in the shoulder-elbow complexes, and discussed optimum configurations in executing different tasks. The results are important for the design and control of humanoid robots, in medicine and sports.

Design of a Bio-Inspired Wearable Exoskeleton for Applications in Robotics In this paper we explain the methodology we adopted to design the kinematics structure of a multi-contact points haptic interface. We based our concept on the analysis of the human arm anatomy and kinematics with the intend to synthesize a system that will be able to interface with the human limb in a very natural way. We proposed a simplified kinematic model of the human arm using a notation coming from the robotics field. To find out the best kinematics architecture we employed real movement data, measured from a human subject, and integrated them with the kinematic model of the exoskeleton, this allow us to test the system before its construction and to formalize specific requirements. We also implemented and tested a first passive version of the shoulder joint.

A Minimal Set Of Coordinates For Describing Humanoid Shoulder Motion The kinematics of the anatomical shoulder are analysed and modelled as a parallel mechanism similar to a Stewart platform. A new method is proposed to describe the shoulder kinematics with minimal coordinates and solve the indeterminacy. The minimal coordinates are defined from bony landmarks and the scapulothoracic kinematic constraints. Independent from one another, they uniquely characterise the shoulder motion. A humanoid mechanism is then proposed with identical kinematic properties. It is then shown how minimal coordinates can be obtained for this mechanism and how the coordinates simplify both the motion-planning task and trajectory-tracking control. Lastly, the coordinates are also shown to have an application in the field of biomechanics where they can be used to model the scapulohumeral rhythm.

Analytical Inverse Kinematics Solver for Anthropomorphic 7-DOF Redundant Manipulators with Human-Like Configuration Constraints. It is a common belief that service robots shall move in a human-like manner to enable natural and convenient interaction with a human user or collaborator. In particular, this applies to anthropomorphic 7-DOF redundant robot manipulators that have a shoulder-elbow-wrist configuration. On the kinematic level, human-like movement then can be realized by means of selecting a redundancy resolution for the inverse kinematics (IK), which realizes human-like movement through respective nullspace preferences. In this paper, key positions are introduced and defined as Cartesian positions of the manipulator's elbow and wrist joints. The key positions are used as constraints on the inverse kinematics in addition to orientation constraints at the end-effector, such that the inverse kinematics can be calculated through an efficient analytical scheme and realizes human-like configurations. To obtain suitable key positions, a correspondence method named wrist-elbow-in-line is derived to map key positions of human demonstrations to the real robot for obtaining a valid analytical inverse kinematics solution. A human demonstration tracking experiment is conducted to evaluate the end-effector accuracy and human-likeness of the generated motion for a 7-DOF Kuka-LWR arm. The results are compared to a similar correspondance method that emphasizes only the wrist postion and show that the subtle differences between the two different correspondence methods may lead to significant performance differences. Furthermore, the wrist-elbow-in-line method is validated as more stable in practical application and extended for obstacle avoidance.

Footprints: history-rich tools for information foraging Inspired by Hill and Hollans original work [7], we have beendeveloping a theory of interaction history and building tools toapply this theory to navigation in a complex information space. Wehave built a series of tools - map, paths, annota- tions andsignposts - based on a physical-world navigation metaphor. Thesetools have been in use for over a year. Our user study involved acontrolled browse task and showed that users were able to get thesame amount of work done with significantly less effort.

Very Deep Convolutional Networks for Large-Scale Image Recognition. In this work we investigate the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting. Our main contribution is a thorough evaluation of networks of increasing depth using an architecture with very small (3x3) convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers. These findings were the basis of our ImageNet Challenge 2014 submission, where our team secured the first and the second places in the localisation and classification tracks respectively. We also show that our representations generalise well to other datasets, where they achieve state-of-the-art results. We have made our two best-performing ConvNet models publicly available to facilitate further research on the use of deep visual representations in computer vision.

Chimp optimization algorithm. •A novel optimizer called Chimp Optimization Algorithm (ChOA) is proposed.•ChOA is inspired by individual intelligence and sexual motivation of chimps.•ChOA alleviates the problems of slow convergence rate and trapping in local optima.•The four main steps of Chimp hunting are implemented.

Space-time modeling of traffic flow. This paper discusses the application of space-time autoregressive integrated moving average (STARIMA) methodology for representing traffic flow patterns. Traffic flow data are in the form of spatial time series and are collected at specific locations at constant intervals of time. Important spatial characteristics of the space-time process are incorporated in the STARIMA model through the use of weighting matrices estimated on the basis of the distances among the various locations where data are collected. These matrices distinguish the space-time approach from the vector autoregressive moving average (VARMA) methodology and enable the model builders to control the number of the parameters that have to be estimated. The proposed models can be used for short-term forecasting of space-time stationary traffic-flow processes and for assessing the impact of traffic-flow changes on other parts of the network. The three-stage iterative space-time model building procedure is illustrated using 7.5min average traffic flow data for a set of 25 loop-detectors located at roads that direct to the centre of the city of Athens, Greece. Data for two months with different traffic-flow characteristics are modelled in order to determine the stability of the parameter estimation.

A novel full structure optimization algorithm for radial basis probabilistic neural networks. In this paper, a novel full structure optimization algorithm for radial basis probabilistic neural networks (RBPNN) is proposed. Firstly, a minimum volume covering hyperspheres (MVCH) algorithm is proposed to heuristically select the initial hidden layer centers of the RBPNN, and then the recursive orthogonal least square (ROLS) algorithm combined with the particle swarm optimization (PSO) algorithm is adopted to further optimize the initial structure of the RBPNN. Finally, the effectiveness and efficiency of our proposed algorithm are evaluated through a plant species identification task involving 50 plant species.

Beamforming for MISO Interference Channels with QoS and RF Energy Transfer We consider a multiuser multiple-input single-output interference channel where the receivers are characterized by both quality-of-service (QoS) and radio-frequency (RF) energy harvesting (EH) constraints. We consider the power splitting RF-EH technique where each receiver divides the received signal into two parts a) for information decoding and b) for battery charging. The minimum required power that supports both the QoS and the RF-EH constraints is formulated as an optimization problem that incorporates the transmitted power and the beamforming design at each transmitter as well as the power splitting ratio at each receiver. We consider both the cases of fixed beamforming and when the beamforming design is incorporated into the optimization problem. For fixed beamforming we study three standard beamforming schemes, the zero-forcing (ZF), the regularized zero-forcing (RZF) and the maximum ratio transmission (MRT); a hybrid scheme, MRT-ZF, comprised of a linear combination of MRT and ZF beamforming is also examined. The optimal solution for ZF beamforming is derived in closed-form, while optimization algorithms based on second-order cone programming are developed for MRT, RZF and MRT-ZF beamforming to solve the problem. In addition, the joint-optimization of beamforming and power allocation is studied using semidefinite programming (SDP) with the aid of rank relaxation.

Scalable and Privacy-Preserving Data Sharing Based on Blockchain. With the development of network technology and cloud computing, data sharing is becoming increasingly popular, and many scholars have conducted in-depth research to promote its flourish. As the scale of data sharing expands, its privacy protection has become a hot issue in research. Moreover, in data sharing, the data is usually maintained in multiple parties, which brings new challenges to protect the privacy of these multi-party data. In this paper, we propose a trusted data sharing scheme using blockchain. We use blockchain to prevent the shared data from being tampered, and use the Paillier cryptosystem to realize the confidentiality of the shared data. In the proposed scheme, the shared data can be traded, and the transaction information is protected by using the (p, t)-threshold Paillier cryptosystem. We conduct experiments in cloud storage scenarios and the experimental results demonstrate the efficiency and effectiveness of the proposed scheme.

Robot tutor and pupils’ educational ability: Teaching the times tables Research shows promising results of educational robots in language and STEM tasks. In language, more research is available, occasionally in view of individual differences in pupils’ educational ability levels, and learning seems to improve with more expressive robot behaviors. In STEM, variations in robots’ behaviors have been examined with inconclusive results and never while systematically investigating how differences in educational abilities match with different robot behaviors. We applied an autonomously tutoring robot (without tablet, partly WOz) in a 2 × 2 experiment of social vs. neutral behavior in above-average vs. below-average schoolchildren (N = 86; age 8–10 years) while rehearsing the multiplication tables on a one-to-one basis. The standard school test showed that on average, pupils significantly improved their performance even after 3 occasions of 5-min exercises. Beyond-average pupils profited most from a robot tutor, whereas those below average in multiplication benefited more from a robot that showed neutral rather than more social behavior.

A three-network architecture for on-line learning and optimization based on adaptive dynamic programming In this paper, we propose a novel adaptive dynamic programming (ADP) architecture with three networks, an action network, a critic network, and a reference network, to develop internal goal-representation for online learning and optimization. Unlike the traditional ADP design normally with an action network and a critic network, our approach integrates the third network, a reference network, into the actor-critic design framework to automatically and adaptively build an internal reinforcement signal to facilitate learning and optimization overtime to accomplish goals. We present the detailed design architecture and its associated learning algorithm to explain how effective learning and optimization can be achieved in this new ADP architecture. Furthermore, we test the performance of our architecture both on the cart-pole balancing task and the triple-link inverted pendulum balancing task, which are the popular benchmarks in the community to demonstrate its learning and control performance over time.

Error bounds of adaptive dynamic programming algorithms for solving undiscounted optimal control problems. In this paper, we establish error bounds of adaptive dynamic programming algorithms for solving undiscounted infinite-horizon optimal control problems of discrete-time deterministic nonlinear systems. We consider approximation errors in the update equations of both value function and control policy. We utilize a new assumption instead of the contraction assumption in discounted optimal control problems. We establish the error bounds for approximate value iteration based on a new error condition. Furthermore, we also establish the error bounds for approximate policy iteration and approximate optimistic policy iteration algorithms. It is shown that the iterative approximate value function can converge to a finite neighborhood of the optimal value function under some conditions. To implement the developed algorithms, critic and action neural networks are used to approximate the value function and control policy, respectively. Finally, a simulation example is given to demonstrate the effectiveness of the developed algorithms.