Datasets:

chreh

/

peer_read_neurips

like 0

EMULATOR vs REAL PHONE: Android Malware Detection Using Machine Learning

The Android operating system has become the most popular operating system for smartphones and tablets leading to a rapid rise in malware. Sophisticated Android malware employ detection avoidance techniques in order to hide their malicious activities from analysis tools. These include a wide range of anti-emulator techniques, where the malware programs attempt to hide their malicious activities by detecting the emulator. For this reason, countermeasures against antiemulation are becoming increasingly important in Android malware detection. Analysis and detection based on real devices can alleviate the problems of anti-emulation as well as improve the effectiveness of dynamic analysis. Hence, in this paper we present an investigation of machine learning based malware detection using dynamic analysis on real devices. A tool is implemented to automatically extract dynamic features from Android phones and through several experiments, a comparative analysis of emulator based vs. device based detection by means of several machine learning algorithms is undertaken. Our study shows that several features could be extracted more effectively from the on-device dynamic analysis compared to emulators. It was also found that approximately 24% more apps were successfully analysed on the phone. Furthermore, all of the studied machine learning based detection performed better when applied to features extracted from the on-device dynamic analysis.

Detecting Overfitting via Adversarial Examples

The repeated community-wide reuse of test sets in popular benchmark problems raises doubts about the credibility of reported test-error rates. Verifying whether a learned model is overfitted to a test set is challenging as independent test sets drawn from the same data distribution are usually unavailable, while other test sets may introduce a distribution shift. We propose a new hypothesis test that uses only the original test data to detect overfitting. It utilizes a new unbiased error estimate that is based on adversarial examples generated from the test data and importance weighting. Overfitting is detected if this error estimate is sufficiently different from the original test error rate. We develop a specialized variant of our test for multiclass image classification, and apply it to testing overfitting of recent models to the popular ImageNet benchmark. Our method correctly indicates overfitting of the trained model to the training set, but is not able to detect any overfitting to the test set, in line with other recent work on this topic.

Application of a Fuzzy Programming Technique to Production Planning in the Textile Industry

Many engineering optimization problems can be considered as linear programming problems where all or some of the parameters involved are linguistic in nature. These can only be quantified using fuzzy sets. The aim of this paper is to solve a fuzzy linear programming problem in which the parameters involved are fuzzy quantities with logistic membership functions. To explore the applicability of the method a numerical example is considered to determine the monthly production planning quotas and profit of a home textile group.

A Consistent Regularization Approach for Structured Prediction

We propose and analyze a regularization approach for structured prediction problems. We characterize a large class of loss functions that allows to naturally embed structured outputs in a linear space. We exploit this fact to design learning algorithms using a surrogate loss approach and regularization techniques. We prove universal consistency and finite sample bounds characterizing the generalization properties of the proposed method. Experimental results are provided to demonstrate the practical usefulness of the proposed approach.

Multi-relational Learning Using Weighted Tensor Decomposition with Modular Loss

We propose a modular framework for multi-relational learning via tensor decomposition. In our learning setting, the training data contains multiple types of relationships among a set of objects, which we represent by a sparse three-mode tensor. The goal is to predict the values of the missing entries. To do so, we model each relationship as a function of a linear combination of latent factors. We learn this latent representation by computing a low-rank tensor decomposition, using quasi-Newton optimization of a weighted objective function. Sparsity in the observed data is captured by the weighted objective, leading to improved accuracy when training data is limited. Exploiting sparsity also improves efficiency, potentially up to an order of magnitude over unweighted approaches. In addition, our framework accommodates arbitrary combinations of smooth, task-specific loss functions, making it better suited for learning different types of relations. For the typical cases of real-valued functions and binary relations, we propose several loss functions and derive the associated parameter gradients. We evaluate our method on synthetic and real data, showing significant improvements in both accuracy and scalability over related factorization techniques.

Comparison of ontology alignment algorithms across single matching task via the McNemar test

Ontology alignment is widely used to find the correspondences between different ontologies in diverse fields. After discovering the alignment by methods, several performance scores are available to evaluate them. The scores require the produced alignment by a method and the reference alignment containing the underlying actual correspondences of the given ontologies. The current trend in alignment evaluation is to put forward a new score and to compare various alignments by juxtaposing their performance scores. However, it is substantially provocative to select one performance score among others for comparison. On top of that, claiming if one method has a better performance than one another can not be substantiated by solely comparing the scores. In this paper, we propose the statistical procedures which enable us to theoretically favor one method over one another. The McNemar test is considered as a reliable and suitable means for comparing two ontology alignment methods over one matching task. The test applies to a 2 x 2 contingency table which can be constructed in two different ways based on the alignments, each of which has their own merits/pitfalls. The ways of the contingency table construction and various apposite statistics from the McNemar test are elaborated in minute detail. In the case of having more than two alignment methods for comparison, the family-wise error rate is expected to happen. Thus, the ways of preventing such an error are also discussed. A directed graph visualizes the outcome of the McNemar test in the presence of multiple alignment methods. From this graph, it is readily understood if one method is better than one another or if their differences are imperceptible. Our investigation on the methods participated in the anatomy track of OAEI 2016 demonstrates that AML and CroMatcher are the top two methods and DKP-AOM and Alin are the bottom two ones.

Nonparametric Bayesian Storyline Detection from Microtexts

News events and social media are composed of evolving storylines, which capture public attention for a limited period of time. Identifying these storylines would enable many high-impact applications, such as tracking public interest and opinion in ongoing crisis events. However, this requires integrating temporal and linguistic information, and prior work takes a largely heuristic approach. We present a novel online non-parametric Bayesian framework for storyline detection, using the distance-dependent Chinese Restaurant Process (dd-CRP). To ensure efficient linear-time inference, we employ a fixed-lag Gibbs sampling procedure, which is novel for the dd-CRP. We evaluate our baseline and proposed models on the TREC Twitter Timeline Generation task and show strong results.

Data Analytics using Ontologies of Management Theories: Towards Implementing 'From Theory to Practice'

We explore how computational ontologies can be impactful vis-a-vis the developing discipline of "data science." We posit an approach wherein management theories are represented as formal axioms, and then applied to draw inferences about data that reside in corporate databases. That is, management theories would be implemented as rules within a data analytics engine. We demonstrate a case study development of such an ontology by formally representing an accounting theory in First-Order Logic. Though quite preliminary, the idea that an information technology, namely ontologies, can potentially actualize the academic cliche, "From Theory to Practice," and be applicable to the burgeoning domain of data analytics is novel and exciting.

A Question Answering Approach to Emotion Cause Extraction

Emotion cause extraction aims to identify the reasons behind a certain emotion expressed in text. It is a much more difficult task compared to emotion classification. Inspired by recent advances in using deep memory networks for question answering (QA), we propose a new approach which considers emotion cause identification as a reading comprehension task in QA. Inspired by convolutional neural networks, we propose a new mechanism to store relevant context in different memory slots to model context information. Our proposed approach can extract both word level sequence features and lexical features. Performance evaluation shows that our method achieves the state-of-the-art performance on a recently released emotion cause dataset, outperforming a number of competitive baselines by at least 3.01% in F-measure.

Improved Deep Learning Baselines for Ubuntu Corpus Dialogs

This paper presents results of our experiments using the Ubuntu Dialog Corpus - the largest publicly available multi-turn dialog corpus. First, we use an in-house implementation of previously reported models to do an independent evaluation using the same data. Second, we evaluate the performances of various LSTMs, Bi-LSTMs and CNNs on the dataset. Third, we create an ensemble by averaging predictions of multiple models. The ensemble further improves the performance and it achieves a state-of-the-art result for this dataset. Finally, we discuss our future plans using this corpus.

Attentive Recurrent Comparators

Attentive Recurrent Comparators (ARCs) are a novel class of neural networks built with attention and recurrence that learn to estimate the similarity of a set of objects by cycling through them and making observations. The observations made in one object are conditioned on the observations made in all the other objects. This allows ARCs to learn to focus on the salient aspects needed to ascertain similarity. Our simplistic model that does not use any convolutions performs comparably to Deep Convolutional Siamese Networks on various visual tasks. However using ARCs and convolutional feature extractors in conjunction produces a model that is significantly better than any other method and has superior generalization capabilities. On the Omniglot dataset, ARC based models achieve an error rate of 1.5\% in the One-Shot classification task - a 2-3x reduction compared to the previous best models. This is also the first Deep Learning model to outperform humans (4.5\%) and surpass the state of the art accuracy set by the highly specialized Hierarchical Bayesian Program Learning (HBPL) system (3.3\%).

Video Frame Synthesis using Deep Voxel Flow

We address the problem of synthesizing new video frames in an existing video, either in-between existing frames (interpolation), or subsequent to them (extrapolation). This problem is challenging because video appearance and motion can be highly complex. Traditional optical-flow-based solutions often fail where flow estimation is challenging, while newer neural-network-based methods that hallucinate pixel values directly often produce blurry results. We combine the advantages of these two methods by training a deep network that learns to synthesize video frames by flowing pixel values from existing ones, which we call deep voxel flow. Our method requires no human supervision, and any video can be used as training data by dropping, and then learning to predict, existing frames. The technique is efficient, and can be applied at any video resolution. We demonstrate that our method produces results that both quantitatively and qualitatively improve upon the state-of-the-art.

Training Dependency Parsers with Partial Annotation

Recently, these has been a surge on studying how to obtain partially annotated data for model supervision. However, there still lacks a systematic study on how to train statistical models with partial annotation (PA). Taking dependency parsing as our case study, this paper describes and compares two straightforward approaches for three mainstream dependency parsers. The first approach is previously proposed to directly train a log-linear graph-based parser (LLGPar) with PA based on a forest-based objective. This work for the first time proposes the second approach to directly training a linear graph-based parse (LGPar) and a linear transition-based parser (LTPar) with PA based on the idea of constrained decoding. We conduct extensive experiments on Penn Treebank under three different settings for simulating PA, i.e., random dependencies, most uncertain dependencies, and dependencies with divergent outputs from the three parsers. The results show that LLGPar is most effective in learning from PA and LTPar lags behind the graph-based counterparts by large margin. Moreover, LGPar and LTPar can achieve best performance by using LLGPar to complete PA into full annotation (FA).

Fast greedy algorithms for dictionary selection with generalized sparsity constraints

In dictionary selection, several atoms are selected from finite candidates that successfully approximate given data points in the sparse representation. We propose a novel efficient greedy algorithm for dictionary selection. Not only does our algorithm work much faster than the known methods, but it can also handle more complex sparsity constraints, such as average sparsity. Using numerical experiments, we show that our algorithm outperforms the known methods for dictionary selection, achieving competitive performances with dictionary learning algorithms in a smaller running time.

Modelling Electricity Consumption in Office Buildings: An Agent Based Approach

In this paper, we develop an agent-based model which integrates four important elements, i.e. organisational energy management policies/regulations, energy management technologies, electric appliances and equipment, and human behaviour, to simulate the electricity consumption in office buildings. Based on a case study, we use this model to test the effectiveness of different electricity management strategies, and solve practical office electricity consumption problems. This paper theoretically contributes to an integration of the four elements involved in the complex organisational issue of office electricity consumption, and practically contributes to an application of an agent-based approach for office building electricity consumption study.

Inverse Density as an Inverse Problem: the Fredholm Equation Approach

In this paper we address the problem of estimating the ratio $\frac{q}{p}$ where $p$ is a density function and $q$ is another density, or, more generally an arbitrary function. Knowing or approximating this ratio is needed in various problems of inference and integration, in particular, when one needs to average a function with respect to one probability distribution, given a sample from another. It is often referred as {\it importance sampling} in statistical inference and is also closely related to the problem of {\it covariate shift} in transfer learning as well as to various MCMC methods. It may also be useful for separating the underlying geometry of a space, say a manifold, from the density function defined on it.

Extended Mixture of MLP Experts by Hybrid of Conjugate Gradient Method and Modified Cuckoo Search

This paper investigates a new method for improving the learning algorithm of Mixture of Experts (ME) model using a hybrid of Modified Cuckoo Search (MCS) and Conjugate Gradient (CG) as a second order optimization technique. The CG technique is combined with Back-Propagation (BP) algorithm to yield a much more efficient learning algorithm for ME structure. In addition, the experts and gating networks in enhanced model are replaced by CG based Multi-Layer Perceptrons (MLPs) to provide faster and more accurate learning. The CG is considerably depends on initial weights of connections of Artificial Neural Network (ANN), so, a metaheuristic algorithm, the so-called Modified Cuckoo Search is applied in order to select the optimal weights. The performance of proposed method is compared with Gradient Decent Based ME (GDME) and Conjugate Gradient Based ME (CGME) in classification and regression problems. The experimental results show that hybrid MSC and CG based ME (MCS-CGME) has faster convergence and better performance in utilized benchmark data sets.

Pruning Random Forests for Prediction on a Budget

We propose to prune a random forest (RF) for resource-constrained prediction. We first construct a RF and then prune it to optimize expected feature cost & accuracy. We pose pruning RFs as a novel 0-1 integer program with linear constraints that encourages feature re-use. We establish total unimodularity of the constraint set to prove that the corresponding LP relaxation solves the original integer program. We then exploit connections to combinatorial optimization and develop an efficient primal-dual algorithm, scalable to large datasets. In contrast to our bottom-up approach, which benefits from good RF initialization, conventional methods are top-down acquiring features based on their utility value and is generally intractable, requiring heuristics. Empirically, our pruning algorithm outperforms existing state-of-the-art resource-constrained algorithms.

Contextual Stochastic Block Models

We provide the first information theoretical tight analysis for inference of latent community structure given a sparse graph along with high dimensional node covariates, correlated with the same latent communities. Our work bridges recent theoretical breakthroughs in detection of latent community structure without nodes covariates and a large body of empirical work using diverse heuristics for combining node covariates with graphs for inference. The tightness of our analysis implies in particular, the information theoretic necessity of combining the different sources of information. Our analysis holds for networks of large degrees as well as for a Gaussian version of the model.

Learning Disentangled Representation for Robust Person Re-identification

We address the problem of person re-identification (reID), that is, retrieving person images from a large dataset, given a query image of the person of interest. The key challenge is to learn person representations robust to intra-class variations, as different persons can have the same attribute and the same person's appearance looks different with viewpoint changes. Recent reID methods focus on learning discriminative features but robust to only a particular factor of variations (e.g., human pose) and this requires corresponding supervisory signals (e.g., pose annotations). To tackle this problem, we propose to disentangle identity-related and -unrelated features from person images. Identity-related features contain information useful for specifying a particular person (e.g.,clothing), while identity-unrelated ones hold other factors (e.g., human pose, scale changes). To this end, we introduce a new generative adversarial network, dubbed identity shuffle GAN (IS-GAN), that factorizes these features using identification labels without any auxiliary information. We also propose an identity shuffling technique to regularize the disentangled features. Experimental results demonstrate the effectiveness of IS-GAN, largely outperforming the state of the art on standard reID benchmarks including the Market-1501, CUHK03 and DukeMTMC-reID. Our code and models will be available online at the time of the publication.

CogALex-V Shared Task: LexNET - Integrated Path-based and Distributional Method for the Identification of Semantic Relations

We present a submission to the CogALex 2016 shared task on the corpus-based identification of semantic relations, using LexNET (Shwartz and Dagan, 2016), an integrated path-based and distributional method for semantic relation classification. The reported results in the shared task bring this submission to the third place on subtask 1 (word relatedness), and the first place on subtask 2 (semantic relation classification), demonstrating the utility of integrating the complementary path-based and distributional information sources in recognizing semantic relatedness. Combined with a common similarity measure, LexNET performs fairly good on the word relatedness task (subtask 1). The relatively lower performance of LexNET and the various other systems on subtask 2, however, confirms the difficulty of the semantic relation classification task, and stresses the need to develop additional methods for this task.

Multiclass Total Variation Clustering

Ideas from the image processing literature have recently motivated a new set of clustering algorithms that rely on the concept of total variation. While these algorithms perform well for bi-partitioning tasks, their recursive extensions yield unimpressive results for multiclass clustering tasks. This paper presents a general framework for multiclass total variation clustering that does not rely on recursion. The results greatly outperform previous total variation algorithms and compare well with state-of-the-art NMF approaches.

Extended Lifted Inference with Joint Formulas

The First-Order Variable Elimination (FOVE) algorithm allows exact inference to be applied directly to probabilistic relational models, and has proven to be vastly superior to the application of standard inference methods on a grounded propositional model. Still, FOVE operators can be applied under restricted conditions, often forcing one to resort to propositional inference. This paper aims to extend the applicability of FOVE by providing two new model conversion operators: the first and the primary is joint formula conversion and the second is just-different counting conversion. These new operations allow efficient inference methods to be applied directly on relational models, where no existing efficient method could be applied hitherto. In addition, aided by these capabilities, we show how to adapt FOVE to provide exact solutions to Maximum Expected Utility (MEU) queries over relational models for decision under uncertainty. Experimental evaluations show our algorithms to provide significant speedup over the alternatives.

Recurrent Neural Network Based Modeling of Gene Regulatory Network Using Bat Algorithm

Correct inference of genetic regulations inside a cell is one of the greatest challenges in post genomic era for the biologist and researchers. Several intelligent techniques and models were already proposed to identify the regulatory relations among genes from the biological database like time series microarray data. Recurrent Neural Network (RNN) is one of the most popular and simple approach to model the dynamics as well as to infer correct dependencies among genes. In this paper, Bat Algorithm (BA) was applied to optimize the model parameters of RNN model of Gene Regulatory Network (GRN). Initially the proposed method is tested against small artificial network without any noise and the efficiency was observed in term of number of iteration, number of population and BA optimization parameters. The model was also validated in presence of different level of random noise for the small artificial network and that proved its ability to infer the correct inferences in presence of noise like real world dataset. In the next phase of this research, BA based RNN is applied to real world benchmark time series microarray dataset of E. Coli. The results shown that it can able to identify the maximum true positive regulation but also include some false positive regulations. Therefore, BA is very suitable for identifying biological plausible GRN with the help RNN model

Forecasting Treatment Responses Over Time Using Recurrent Marginal Structural Networks

Electronic health records provide a rich source of data for machine learning methods to learn dynamic treatment responses over time. However, any direct estimation is hampered by the presence of time-dependent confounding, where actions taken are dependent on time-varying variables related to the outcome of interest. Drawing inspiration from marginal structural models, a class of methods in epidemiology which use propensity weighting to adjust for time-dependent confounders, we introduce the Recurrent Marginal Structural Network - a sequence-to-sequence architecture for forecasting a patient's expected response to a series of planned treatments. Using simulations of a state-of-the-art pharmacokinetic-pharmacodynamic (PK-PD) model of tumor growth, we demonstrate the ability of our network to accurately learn unbiased treatment responses from observational data – even under changes in the policy of treatment assignments – and performance gains over benchmarks.

R$^3$: Reinforced Reader-Ranker for Open-Domain Question Answering

In recent years researchers have achieved considerable success applying neural network methods to question answering (QA). These approaches have achieved state of the art results in simplified closed-domain settings such as the SQuAD (Rajpurkar et al., 2016) dataset, which provides a pre-selected passage, from which the answer to a given question may be extracted. More recently, researchers have begun to tackle open-domain QA, in which the model is given a question and access to a large corpus (e.g., wikipedia) instead of a pre-selected passage (Chen et al., 2017a). This setting is more complex as it requires large-scale search for relevant passages by an information retrieval component, combined with a reading comprehension model that "reads" the passages to generate an answer to the question. Performance in this setting lags considerably behind closed-domain performance.

b-bit Marginal Regression

We consider the problem of sparse signal recovery from $m$ linear measurements quantized to $b$ bits. $b$-bit Marginal Regression is proposed as recovery algorithm. We study the question of choosing $b$ in the setting of a given budget of bits $B = m \cdot b$ and derive a single easy-to-compute expression characterizing the trade-off between $m$ and $b$. The choice $b = 1$ turns out to be optimal for estimating the unit vector corresponding to the signal for any level of additive Gaussian noise before quantization as well as for adversarial noise. For $b \geq 2$, we show that Lloyd-Max quantization constitutes an optimal quantization scheme and that the norm of the signal canbe estimated consistently by maximum likelihood.

Fast Proximal Linearized Alternating Direction Method of Multiplier with Parallel Splitting

The Augmented Lagragian Method (ALM) and Alternating Direction Method of Multiplier (ADMM) have been powerful optimization methods for general convex programming subject to linear constraint. We consider the convex problem whose objective consists of a smooth part and a nonsmooth but simple part. We propose the Fast Proximal Augmented Lagragian Method (Fast PALM) which achieves the convergence rate $O(1/K^2)$, compared with $O(1/K)$ by the traditional PALM. In order to further reduce the per-iteration complexity and handle the multi-blocks problem, we propose the Fast Proximal ADMM with Parallel Splitting (Fast PL-ADMM-PS) method. It also partially improves the rate related to the smooth part of the objective function. Experimental results on both synthesized and real world data demonstrate that our fast methods significantly improve the previous PALM and ADMM.

Decision Tree Classification on Outsourced Data

This paper proposes a client-server decision tree learning method for outsourced private data. The privacy model is anatomization/fragmentation: the server sees data values, but the link between sensitive and identifying information is encrypted with a key known only to clients. Clients have limited processing and storage capability. Both sensitive and identifying information thus are stored on the server. The approach presented also retains most processing at the server, and client-side processing is amortized over predictions made by the clients. Experiments on various datasets show that the method produces decision trees approaching the accuracy of a non-private decision tree, while substantially reducing the client's computing resource requirements.

A Robust Transformation-Based Learning Approach Using Ripple Down Rules for Part-of-Speech Tagging

This paper presents a new method to construct a system of transformation rules for the Part-Of-Speech tagging task. Our approach is based on an incremental knowledge acquisition methodology where rules are stored in an exception-structure and new rules are only added to correct errors of existing rules; thus allowing systematic control of the interaction between rules. Experiments on 13 languages exhibit that our method is fast in terms of training time and tagging speed. Furthermore, our method is able to attain state-of-the-art accuracies for relatively isolating or analytic languages whilst reaching competitive accuracy results on morphologically rich Indo-European languages.

Asymmetric Tri-training for Unsupervised Domain Adaptation

Deep-layered models trained on a large number of labeled samples boost the accuracy of many tasks. It is important to apply such models to different domains because collecting many labeled samples in various domains is expensive. In unsupervised domain adaptation, one needs to train a classifier that works well on a target domain when provided with labeled source samples and unlabeled target samples. Although many methods aim to match the distributions of source and target samples, simply matching the distribution cannot ensure accuracy on the target domain. To learn discriminative representations for the target domain, we assume that artificially labeling target samples can result in a good representation. Tri-training leverages three classifiers equally to give pseudo-labels to unlabeled samples, but the method does not assume labeling samples generated from a different domain.In this paper, we propose an asymmetric tri-training method for unsupervised domain adaptation, where we assign pseudo-labels to unlabeled samples and train neural networks as if they are true labels. In our work, we use three networks asymmetrically. By asymmetric, we mean that two networks are used to label unlabeled target samples and one network is trained by the samples to obtain target-discriminative representations. We evaluate our method on digit recognition and sentiment analysis datasets. Our proposed method achieves state-of-the-art performance on the benchmark digit recognition datasets of domain adaptation.

Trolls Identification within an Uncertain Framework

The web plays an important role in people's social lives since the emergence of Web 2.0. It facilitates the interaction between users, gives them the possibility to freely interact, share and collaborate through social networks, online communities forums, blogs, wikis and other online collaborative media. However, an other side of the web is negatively taken such as posting inflammatory messages. Thus, when dealing with the online communities forums, the managers seek to always enhance the performance of such platforms. In fact, to keep the serenity and prohibit the disturbance of the normal atmosphere, managers always try to novice users against these malicious persons by posting such message (DO NOT FEED TROLLS). But, this kind of warning is not enough to reduce this phenomenon. In this context we propose a new approach for detecting malicious people also called 'Trolls' in order to allow community managers to take their ability to post online. To be more realistic, our proposal is defined within an uncertain framework. Based on the assumption consisting on the trolls' integration in the successful discussion threads, we try to detect the presence of such malicious users. Indeed, this method is based on a conflict measure of the belief function theory applied between the different messages of the thread. In order to show the feasibility and the result of our approach, we test it in different simulated data.

Causal Discovery for Manufacturing Domains

Yield and quality improvement is of paramount importance to any manufacturing company. One of the ways of improving yield is through discovery of the root causal factors affecting yield. We propose the use of data-driven interpretable causal models to identify key factors affecting yield. We focus on factors that are measured in different stages of production and testing in the manufacturing cycle of a product. We apply causal structure learning techniques on real data collected from this line. Specifically, the goal of this work is to learn interpretable causal models from observational data produced by manufacturing lines.

Bounded Regret for Finite-Armed Structured Bandits

We study a new type of K-armed bandit problem where the expected return of one arm may depend on the returns of other arms. We present a new algorithm for this general class of problems and show that under certain circumstances it is possible to achieve finite expected cumulative regret. We also give problem-dependent lower bounds on the cumulative regret showing that at least in special cases the new algorithm is nearly optimal.

Structured Prediction Theory Based on Factor Graph Complexity

We present a general theoretical analysis of structured prediction with a series of new results. We give new data-dependent margin guarantees for structured prediction for a very wide family of loss functions and a general family of hypotheses, with an arbitrary factor graph decomposition. These are the tightest margin bounds known for both standard multi-class and general structured prediction problems. Our guarantees are expressed in terms of a data-dependent complexity measure, \emph{factor graph complexity}, which we show can be estimated from data and bounded in terms of familiar quantities for several commonly used hypothesis sets, and a sparsity measure for features and graphs. Our proof techniques include generalizations of Talagrand's contraction lemma that can be of independent interest. We further extend our theory by leveraging the principle of Voted Risk Minimization (VRM) and show that learning is possible even with complex factor graphs. We present new learning bounds for this advanced setting, which we use to devise two new algorithms, \emph{Voted Conditional Random Field} (VCRF) and \emph{Voted Structured Boosting} (StructBoost). These algorithms can make use of complex features and factor graphs and yet benefit from favorable learning guarantees. We also report the results of experiments with VCRF on several datasets to validate our theory.

Optimal Neural Codes for Control and Estimation

Agents acting in the natural world aim at selecting appropriate actions based on noisy and partial sensory observations. Many behaviors leading to decision making and action selection in a closed loop setting are naturally phrased within a control theoretic framework. Within the framework of optimal Control Theory, one is usually given a cost function which is minimized by selecting a control law based on the observations. While in standard control settings the sensors are assumed fixed, biological systems often gain from the extra flexibility of optimizing the sensors themselves. However, this sensory adaptation is geared towards control rather than perception, as is often assumed. In this work we show that sensory adaptation for control differs from sensory adaptation for perception, even for simple control setups. This implies, consistently with recent experimental results, that when studying sensory adaptation, it is essential to account for the task being performed.

A scaled Bregman theorem with applications

Bregman divergences play a central role in the design and analysis of a range of machine learning algorithms. This paper explores the use of Bregman divergences to establish reductions between such algorithms and their analyses. We present a new scaled isodistortion theorem involving Bregman divergences (scaled Bregman theorem for short) which shows that certain "Bregman distortions'" (employing a potentially non-convex generator) may be exactly re-written as a scaled Bregman divergence computed over transformed data. Admissible distortions include geodesic distances on curved manifolds and projections or gauge-normalisation, while admissible data include scalars, vectors and matrices.

Semi-flat minima and saddle points by embedding neural networks to overparameterization

We theoretically study the landscape of the training error for neural networks in overparameterized cases. We consider three basic methods for embedding a network into a wider one with more hidden units, and discuss whether a minimum point of the narrower network gives a minimum or saddle point of the wider one. Our results show that the networks with smooth and ReLU activation have different partially flat landscapes around the embedded point. We also relate these results to a difference of their generalization abilities in overparameterized realization.

A Preliminary Study on the Learning Informativeness of Data Subsets

Estimating the internal state of a robotic system is complex: this is performed from multiple heterogeneous sensor inputs and knowledge sources. Discretization of such inputs is done to capture saliences, represented as symbolic information, which often presents structure and recurrence. As these sequences are used to reason over complex scenarios, a more compact representation would aid exactness of technical cognitive reasoning capabilities, which are today constrained by computational complexity issues and fallback to representational heuristics or human intervention. Such problems need to be addressed to ensure timely and meaningful human-robot interaction. Our work is towards understanding the variability of learning informativeness when training on subsets of a given input dataset. This is in view of reducing the training size while retaining the majority of the symbolic learning potential. We prove the concept on human-written texts, and conjecture this work will reduce training data size of sequential instructions, while preserving semantic relations, when gathering information from large remote sources.

Unsupervised Risk Estimation Using Only Conditional Independence Structure

We show how to estimate a model's test error from unlabeled data, on distributions very different from the training distribution, while assuming only that certain conditional independencies are preserved between train and test. We do not need to assume that the optimal predictor is the same between train and test, or that the true distribution lies in any parametric family. We can also efficiently differentiate the error estimate to perform unsupervised discriminative learning. Our technical tool is the method of moments, which allows us to exploit conditional independencies in the absence of a fully-specified model. Our framework encompasses a large family of losses including the log and exponential loss, and extends to structured output settings such as hidden Markov models.

Une mesure d'expertise pour le crowdsourcing

Crowdsourcing, a major economic issue, is the fact that the firm outsources internal task to the crowd. It is a form of digital subcontracting for the general public. The evaluation of the participants work quality is a major issue in crowdsourcing. Indeed, contributions must be controlled to ensure the effectiveness and relevance of the campaign. We are particularly interested in small, fast and not automatable tasks. Several methods have been proposed to solve this problem, but they are applicable when the "golden truth" is not always known. This work has the particularity to propose a method for calculating the degree of expertise in the presence of gold data in crowdsourcing. This method is based on the belief function theory and proposes a structuring of data using graphs. The proposed approach will be assessed and applied to the data.

Geometry of Compositionality

This paper proposes a simple test for compositionality (i.e., literal usage) of a word or phrase in a context-specific way. The test is computationally simple, relying on no external resources and only uses a set of trained word vectors. Experiments show that the proposed method is competitive with state of the art and displays high accuracy in context-specific compositionality detection of a variety of natural language phenomena (idiomaticity, sarcasm, metaphor) for different datasets in multiple languages. The key insight is to connect compositionality to a curious geometric property of word embeddings, which is of independent interest.

The Role of Normalization in the Belief Propagation Algorithm

An important part of problems in statistical physics and computer science can be expressed as the computation of marginal probabilities over a Markov Random Field. The belief propagation algorithm, which is an exact procedure to compute these marginals when the underlying graph is a tree, has gained its popularity as an efficient way to approximate them in the more general case. In this paper, we focus on an aspect of the algorithm that did not get that much attention in the literature, which is the effect of the normalization of the messages. We show in particular that, for a large class of normalization strategies, it is possible to focus only on belief convergence. Following this, we express the necessary and sufficient conditions for local stability of a fixed point in terms of the graph structure and the beliefs values at the fixed point. We also explicit some connexion between the normalization constants and the underlying Bethe Free Energy.

fastFM: A Library for Factorization Machines

Factorization Machines (FM) are only used in a narrow range of applications and are not part of the standard toolbox of machine learning models. This is a pity, because even though FMs are recognized as being very successful for recommender system type applications they are a general model to deal with sparse and high dimensional features. Our Factorization Machine implementation provides easy access to many solvers and supports regression, classification and ranking tasks. Such an implementation simplifies the use of FM's for a wide field of applications. This implementation has the potential to improve our understanding of the FM model and drive new development.

Convex Calibrated Surrogates for Low-Rank Loss Matrices with Applications to Subset Ranking Losses

The design of convex, calibrated surrogate losses, whose minimization entails consistency with respect to a desired target loss, is an important concept to have emerged in the theory of machine learning in recent years. We give an explicit construction of a convex least-squares type surrogate loss that can be designed to be calibrated for any multiclass learning problem for which the target loss matrix has a low-rank structure; the surrogate loss operates on a surrogate target space of dimension at most the rank of the target loss. We use this result to design convex calibrated surrogates for a variety of subset ranking problems, with target losses including the precision@q, expected rank utility, mean average precision, and pairwise disagreement.

An efficient algorithm for learning with semi-bandit feedback

We consider the problem of online combinatorial optimization under semi-bandit feedback. The goal of the learner is to sequentially select its actions from a combinatorial decision set so as to minimize its cumulative loss. We propose a learning algorithm for this problem based on combining the Follow-the-Perturbed-Leader (FPL) prediction method with a novel loss estimation procedure called Geometric Resampling (GR). Contrary to previous solutions, the resulting algorithm can be efficiently implemented for any decision set where efficient offline combinatorial optimization is possible at all. Assuming that the elements of the decision set can be described with d-dimensional binary vectors with at most m non-zero entries, we show that the expected regret of our algorithm after T rounds is O(m sqrt(dT log d)). As a side result, we also improve the best known regret bounds for FPL in the full information setting to O(m^(3/2) sqrt(T log d)), gaining a factor of sqrt(d/m) over previous bounds for this algorithm.

A System for Probabilistic Linking of Thesauri and Classification Systems

This paper presents a system which creates and visualizes probabilistic semantic links between concepts in a thesaurus and classes in a classification system. For creating the links, we build on the Polylingual Labeled Topic Model (PLL-TM). PLL-TM identifies probable thesaurus descriptors for each class in the classification system by using information from the natural language text of documents, their assigned thesaurus descriptors and their designated classes. The links are then presented to users of the system in an interactive visualization, providing them with an automatically generated overview of the relations between the thesaurus and the classification system.

Divide-and-Conquer based Ensemble to Spot Emotions in Speech using MFCC and Random Forest

Besides spoken words, speech signals also carry information about speaker gender, age, and emotional state which can be used in a variety of speech analysis applications. In this paper, a divide and conquer strategy for ensemble classification has been proposed to recognize emotions in speech. Intrinsic hierarchy in emotions has been utilized to construct an emotions tree, which assisted in breaking down the emotion recognition task into smaller sub tasks. The proposed framework generates predictions in three phases. Firstly, emotions are detected in the input speech signal by classifying it as neutral or emotional. If the speech is classified as emotional, then in the second phase, it is further classified into positive and negative classes. Finally, individual positive or negative emotions are identified based on the outcomes of the previous stages. Several experiments have been performed on a widely used benchmark dataset. The proposed method was able to achieve improved recognition rates as compared to several other approaches.

Lifelong Learning with Non-i.i.d. Tasks

In this work we aim at extending theoretical foundations of lifelong learning. Previous work analyzing this scenario is based on the assumption that the tasks are sampled i.i.d. from a task environment or limited to strongly constrained data distributions. Instead we study two scenarios when lifelong learning is possible, even though the observed tasks do not form an i.i.d. sample: first, when they are sampled from the same environment, but possibly with dependencies, and second, when the task environment is allowed to change over time. In the first case we prove a PAC-Bayesian theorem, which can be seen as a direct generalization of the analogous previous result for the i.i.d. case. For the second scenario we propose to learn an inductive bias in form of a transfer procedure. We present a generalization bound and show on a toy example how it can be used to identify a beneficial transfer algorithm.

Neural Networks and Rational Functions

Neural networks and rational functions efficiently approximate each other. In more detail, it is shown here that for any ReLU network, there exists a rational function of degree $O(\text{polylog}(1/\epsilon))$ which is $\epsilon$-close, and similarly for any rational function there exists a ReLU network of size $O(\text{polylog}(1/\epsilon))$ which is $\epsilon$-close. By contrast, polynomials need degree $\Omega(\text{poly}(1/\epsilon))$ to approximate even a single ReLU. When converting a ReLU network to a rational function as above, the hidden constants depend exponentially on the number of layers, which is shown to be tight; in other words, a compositional representation can be beneficial even for rational functions.

Unsupervised Learning of Word-Sequence Representations from Scratch via Convolutional Tensor Decomposition

Text embeddings have played a key role in obtaining state-of-the-art results in natural language processing. Word2Vec and its variants have successfully mapped words with similar syntactic or semantic meanings to nearby vectors. However, extracting universal embeddings of longer word-sequences remains a challenging task. We employ the convolutional dictionary model for unsupervised learning of embeddings for variable length word-sequences. We propose a two-phase ConvDic+DeconvDec framework that first learns dictionary elements (i.e., phrase templates), and then employs them for decoding the activations. The estimated activations are then used as embeddings for downstream tasks such as sentiment analysis, paraphrase detection, and semantic textual similarity estimation. We propose a convolutional tensor decomposition algorithm for learning the phrase templates. It is shown to be more accurate, and much more efficient than the popular alternating minimization in dictionary learning literature. Our word-sequence embeddings achieve state-of-the-art performance in sentiment classification, semantic textual similarity estimation, and paraphrase detection over eight datasets from various domains, without requiring pre-training or additional features.

Using Dimension Reduction to Improve the Classification of High-dimensional Data

In this work we show that the classification performance of high-dimensional structural MRI data with only a small set of training examples is improved by the usage of dimension reduction methods. We assessed two different dimension reduction variants: feature selection by ANOVA F-test and feature transformation by PCA. On the reduced datasets, we applied common learning algorithms using 5-fold cross-validation. Training, tuning of the hyperparameters, as well as the performance evaluation of the classifiers was conducted using two different performance measures: Accuracy, and Receiver Operating Characteristic curve (AUC). Our hypothesis is supported by experimental results.

Political Speech Generation

In this report we present a system that can generate political speeches for a desired political party. Furthermore, the system allows to specify whether a speech should hold a supportive or opposing opinion. The system relies on a combination of several state-of-the-art NLP methods which are discussed in this report. These include n-grams, Justeson & Katz POS tag filter, recurrent neural networks, and latent Dirichlet allocation. Sequences of words are generated based on probabilities obtained from two underlying models: A language model takes care of the grammatical correctness while a topic model aims for textual consistency. Both models were trained on the Convote dataset which contains transcripts from US congressional floor debates. Furthermore, we present a manual and an automated approach to evaluate the quality of generated speeches. In an experimental evaluation generated speeches have shown very high quality in terms of grammatical correctness and sentence transitions.

Machine learning on images using a string-distance

We present a new method for image feature-extraction which is based on representing an image by a finite-dimensional vector of distances that measure how different the image is from a set of image prototypes. We use the recently introduced Universal Image Distance (UID) \cite{RatsabyChesterIEEE2012} to compare the similarity between an image and a prototype image. The advantage in using the UID is the fact that no domain knowledge nor any image analysis need to be done. Each image is represented by a finite dimensional feature vector whose components are the UID values between the image and a finite set of image prototypes from each of the feature categories. The method is automatic since once the user selects the prototype images, the feature vectors are automatically calculated without the need to do any image analysis. The prototype images can be of different size, in particular, different than the image size. Based on a collection of such cases any supervised or unsupervised learning algorithm can be used to train and produce an image classifier or image cluster analysis. In this paper we present the image feature-extraction method and use it on several supervised and unsupervised learning experiments for satellite image data.

Interest-Driven Discovery of Local Process Models

Local Process Models (LPM) describe structured fragments of process behavior occurring in the context of less structured business processes. Traditional LPM discovery aims to generate a collection of process models that describe highly frequent behavior, but these models do not always provide useful answers for questions posed by process analysts aiming at business process improvement. We propose a framework for goal-driven LPM discovery, based on utility functions and constraints. We describe four scopes on which these utility functions and constrains can be defined, and show that utility functions and constraints on different scopes can be combined to form composite utility functions/constraints. Finally, we demonstrate the applicability of our approach by presenting several actionable business insights discovered with LPM discovery on two real life data sets.

A Diversity-Promoting Objective Function for Neural Conversation Models

Sequence-to-sequence neural network models for generation of conversational responses tend to generate safe, commonplace responses (e.g., \textit{I don't know}) regardless of the input. We suggest that the traditional objective function, i.e., the likelihood of output (responses) given input (messages) is unsuited to response generation tasks. Instead we propose using Maximum Mutual Information (MMI) as objective function in neural models. Experimental results demonstrate that the proposed objective function produces more diverse, interesting, and appropriate responses, yielding substantive gains in \bleu scores on two conversational datasets.

Mining GOLD Samples for Conditional GANs

Conditional generative adversarial networks (cGANs) have gained a considerable attention in recent years due to its class-wise controllability and superior quality for complex generation tasks. We introduce a simple yet effective approach to improving cGANs by measuring the discrepancy between the data distribution and the model distribution on given samples. The proposed measure, coined the gap of log-densities (GOLD), provides an effective self-diagnosis for cGANs while being efficiently, computed from the discriminator. We propose three applications of the GOLD: example re-weighting, rejection sampling, and active learning, which improve the training, inference, and data selection of cGANs, respectively. Our experimental results demonstrate that the proposed methods outperform corresponding baselines for all three applications on different image datasets.

Ontology Based SMS Controller for Smart Phones

Text analysis includes lexical analysis of the text and has been widely studied and used in diverse applications. In the last decade, researchers have proposed many efficient solutions to analyze / classify large text dataset, however, analysis / classification of short text is still a challenge because 1) the data is very sparse 2) It contains noise words and 3) It is difficult to understand the syntactical structure of the text. Short Messaging Service (SMS) is a text messaging service for mobile/smart phone and this service is frequently used by all mobile users. Because of the popularity of SMS service, marketing companies nowadays are also using this service for direct marketing also known as SMS marketing.In this paper, we have proposed Ontology based SMS Controller which analyze the text message and classify it using ontology aslegitimate or spam. The proposed system has been tested on different scenarios and experimental results shows that the proposed solution is effective both in terms of efficiency and time.

Linear Regression with Limited Observation

We consider the most common variants of linear regression, including Ridge, Lasso and Support-vector regression, in a setting where the learner is allowed to observe only a fixed number of attributes of each example at training time. We present simple and efficient algorithms for these problems: for Lasso and Ridge regression they need the same total number of attributes (up to constants) as do full-information algorithms, for reaching a certain accuracy. For Support-vector regression, we require exponentially less attributes compared to the state of the art. By that, we resolve an open problem recently posed by Cesa-Bianchi et al. (2010). Experiments show the theoretical bounds to be justified by superior performance compared to the state of the art.

Partitioning Structure Learning for Segmented Linear Regression Trees

This paper proposes a partitioning structure learning method for segmented linear regression trees (SLRT), which assigns linear predictors over the terminal nodes. The recursive partitioning process is driven by an adaptive split selection algorithm that maximizes, at each node, a criterion function based on a conditional Kendall’s τ statistic that measures the rank dependence between the regressors and the fit- ted linear residuals. Theoretical analysis shows that the split selection algorithm permits consistent identification and estimation of the unknown segments. A suffi- ciently large tree is induced by applying the split selection algorithm recursively. Then the minimal cost-complexity tree pruning procedure is applied to attain the right-sized tree, that ensures (i) the nested structure of pruned subtrees and (ii) consistent estimation to the number of segments. Implanting the SLRT as the built-in base predictor, we obtain the ensemble predictors by random forests (RF) and the proposed weighted random forests (WRF). The practical performance of the SLRT and its ensemble versions are evaluated via numerical simulations and empirical studies. The latter shows their advantageous predictive performance over a set of state-of-the-art tree-based models on well-studied public datasets.

Identifying and Categorizing Anomalies in Retinal Imaging Data

The identification and quantification of markers in medical images is critical for diagnosis, prognosis and management of patients in clinical practice. Supervised- or weakly supervised training enables the detection of findings that are known a priori. It does not scale well, and a priori definition limits the vocabulary of markers to known entities reducing the accuracy of diagnosis and prognosis. Here, we propose the identification of anomalies in large-scale medical imaging data using healthy examples as a reference. We detect and categorize candidates for anomaly findings untypical for the observed data. A deep convolutional autoencoder is trained on healthy retinal images. The learned model generates a new feature representation, and the distribution of healthy retinal patches is estimated by a one-class support vector machine. Results demonstrate that we can identify pathologic regions in images without using expert annotations. A subsequent clustering categorizes findings into clinically meaningful classes. In addition the learned features outperform standard embedding approaches in a classification task.

TSEB: More Efficient Thompson Sampling for Policy Learning

In model-based solution approaches to the problem of learning in an unknown environment, exploring to learn the model parameters takes a toll on the regret. The optimal performance with respect to regret or PAC bounds is achievable, if the algorithm exploits with respect to reward or explores with respect to the model parameters, respectively. In this paper, we propose TSEB, a Thompson Sampling based algorithm with adaptive exploration bonus that aims to solve the problem with tighter PAC guarantees, while being cautious on the regret as well. The proposed approach maintains distributions over the model parameters which are successively refined with more experience. At any given time, the agent solves a model sampled from this distribution, and the sampled reward distribution is skewed by an exploration bonus in order to generate more informative exploration. The policy by solving is then used for generating more experience that helps in updating the posterior over the model parameters. We provide a detailed analysis of the PAC guarantees, and convergence of the proposed approach. We show that our adaptive exploration bonus encourages the additional exploration required for better PAC bounds on the algorithm. We provide empirical analysis on two different simulated domains.

Computational Algorithms Based on the Paninian System to Process Euphonic Conjunctions for Word Searches

Searching for words in Sanskrit E-text is a problem that is accompanied by complexities introduced by features of Sanskrit such as euphonic conjunctions or sandhis. A word could occur in an E-text in a transformed form owing to the operation of rules of sandhi. Simple word search would not yield these transformed forms of the word. Further, there is no search engine in the literature that can comprehensively search for words in Sanskrit E-texts taking euphonic conjunctions into account. This work presents an optimal binary representational schema for letters of the Sanskrit alphabet along with algorithms to efficiently process the sandhi rules of Sanskrit grammar. The work further presents an algorithm that uses the sandhi processing algorithm to perform a comprehensive word search on E-text.

Identifying Bengali Multiword Expressions using Semantic Clustering

One of the key issues in both natural language understanding and generation is the appropriate processing of Multiword Expressions (MWEs). MWEs pose a huge problem to the precise language processing due to their idiosyncratic nature and diversity in lexical, syntactical and semantic properties. The semantics of a MWE cannot be expressed after combining the semantics of its constituents. Therefore, the formalism of semantic clustering is often viewed as an instrument for extracting MWEs especially for resource constraint languages like Bengali. The present semantic clustering approach contributes to locate clusters of the synonymous noun tokens present in the document. These clusters in turn help measure the similarity between the constituent words of a potentially candidate phrase using a vector space model and judge the suitability of this phrase to be a MWE. In this experiment, we apply the semantic clustering approach for noun-noun bigram MWEs, though it can be extended to any types of MWEs. In parallel, the well known statistical models, namely Point-wise Mutual Information (PMI), Log Likelihood Ratio (LLR), Significance function are also employed to extract MWEs from the Bengali corpus. The comparative evaluation shows that the semantic clustering approach outperforms all other competing statistical models. As a by-product of this experiment, we have started developing a standard lexicon in Bengali that serves as a productive Bengali linguistic thesaurus.

One-vs-Each Approximation to Softmax for Scalable Estimation of Probabilities

The softmax representation of probabilities for categorical variables plays a prominent role in modern machine learning with numerous applications in areas such as large scale classification, neural language modeling and recommendation systems. However, softmax estimation is very expensive for large scale inference because of the high cost associated with computing the normalizing constant. Here, we introduce an efficient approximation to softmax probabilities which takes the form of a rigorous lower bound on the exact probability. This bound is expressed as a product over pairwise probabilities and it leads to scalable estimation based on stochastic optimization. It allows us to perform doubly stochastic estimation by subsampling both training instances and class labels. We show that the new bound has interesting theoretical properties and we demonstrate its use in classification problems.

Compositional Sequence Labeling Models for Error Detection in Learner Writing

In this paper, we present the first experiments using neural network models for the task of error detection in learner writing. We perform a systematic comparison of alternative compositional architectures and propose a framework for error detection based on bidirectional LSTMs. Experiments on the CoNLL-14 shared task dataset show the model is able to outperform other participants on detecting errors in learner writing. Finally, the model is integrated with a publicly deployed self-assessment system, leading to performance comparable to human annotators.

An Evaluation of Information Sharing Parking Guidance Policies Using a Bayesian Approach

Real-time parking occupancy information is critical for a parking management system to facilitate drivers to park more efficiently. Recent advances in connected and automated vehicle technologies enable sensor-equipped cars (probe cars) to detect and broadcast available parking spaces when driving through parking lots. In this paper, we evaluate the impact of market penetration of probe cars on the system performance, and investigate different parking guidance policies to improve the data acquisition process. We adopt a simulation-based approach to impose four policies on an off- street parking lot influencing the behavior of probe cars to park in assigned parking spaces. This in turn effects the scanning route and the parking space occupancy estimations. The last policy we propose is a near-optimal guidance strategy that maximizes the information gain of posteriors. The results suggest that an efficient information gathering policy can compensate for low penetration of connected and automated vehicles. We also highlight the policy trade-off that occur while attempting to maximize information gain through explorations and improve assignment accuracy through exploitations. Our results can assist urban policy makers in designing and managing smart parking systems.

On the "Calligraphy" of Books

Authorship attribution is a natural language processing task that has been widely studied, often by considering small order statistics. In this paper, we explore a complex network approach to assign the authorship of texts based on their mesoscopic representation, in an attempt to capture the flow of the narrative. Indeed, as reported in this work, such an approach allowed the identification of the dominant narrative structure of the studied authors. This has been achieved due to the ability of the mesoscopic approach to take into account relationships between different, not necessarily adjacent, parts of the text, which is able to capture the story flow. The potential of the proposed approach has been illustrated through principal component analysis, a comparison with the chance baseline method, and network visualization. Such visualizations reveal individual characteristics of the authors, which can be understood as a kind of calligraphy.

Breaking Symmetries in Graph Search with Canonizing Sets

There are many complex combinatorial problems which involve searching for an undirected graph satisfying given constraints. Such problems are often highly challenging because of the large number of isomorphic representations of their solutions. This paper introduces effective and compact, complete symmetry breaking constraints for small graph search. Enumerating with these symmetry breaks generates all and only non-isomorphic solutions. For small search problems, with up to $10$ vertices, we compute instance independent symmetry breaking constraints. For small search problems with a larger number of vertices we demonstrate the computation of instance dependent constraints which are complete. We illustrate the application of complete symmetry breaking constraints to extend two known sequences from the OEIS related to graph numeration.

BinGAN: Learning Compact Binary Descriptors with a Regularized GAN

In this paper, we propose a novel regularization method for Generative Adversarial Networks that allows the model to learn discriminative yet compact binary representations of image patches (image descriptors). We exploit the dimensionality reduction that takes place in the intermediate layers of the discriminator network and train the binarized penultimate layer's low-dimensional representation to mimic the distribution of the higher-dimensional preceding layers. To achieve this, we introduce two loss terms that aim at: (i) reducing the correlation between the dimensions of the binarized penultimate layer's low-dimensional representation (i.e. maximizing joint entropy) and (ii) propagating the relations between the dimensions in the high-dimensional space to the low-dimensional space. We evaluate the resulting binary image descriptors on two challenging applications, image matching and retrieval, where they achieve state-of-the-art results.

Recognizing Implicit Discourse Relations via Repeated Reading: Neural Networks with Multi-Level Attention

Recognizing implicit discourse relations is a challenging but important task in the field of Natural Language Processing. For such a complex text processing task, different from previous studies, we argue that it is necessary to repeatedly read the arguments and dynamically exploit the efficient features useful for recognizing discourse relations. To mimic the repeated reading strategy, we propose the neural networks with multi-level attention (NNMA), combining the attention mechanism and external memories to gradually fix the attention on some specific words helpful to judging the discourse relations. Experiments on the PDTB dataset show that our proposed method achieves the state-of-art results. The visualization of the attention weights also illustrates the progress that our model observes the arguments on each level and progressively locates the important words.

An n-ary Constraint for the Stable Marriage Problem

We present an n-ary constraint for the stable marriage problem. This constraint acts between two sets of integer variables where the domains of those variables represent preferences. Our constraint enforces stability and disallows bigamy. For a stable marriage instance with $n$ men and $n$ women we require only one of these constraints, and the complexity of enforcing arc-consistency is $O(n^2)$ which is optimal in the size of input. Our computational studies show that our n-ary constraint is significantly faster and more space efficient than the encodings presented in \cite{cp01}. We also introduce a new problem to the constraint community, the sex-equal stable marriage problem.

Deep Action Sequence Learning for Causal Shape Transformation

Deep learning (DL) became the method of choice in recent years for solving problems ranging from object recognition and speech recognition to robotic perception and human disease prediction. In this paper, we present a hybrid architecture of convolutional neural networks (CNN) and stacked autoencoders (SAE) to learn a sequence of actions that nonlinearly transforms an input shape or distribution into a target shape or distribution with the same support. While such a framework can be useful in a variety of problems such as robotic path planning, sequential decision-making in games and identifying material processing pathways to achieve desired microstructures, this paper focuses on controlling fluid deformations in a microfluidic channel by deliberately placing a sequence of pillars, which has a significant impact on manufacturing for biomedical and textile applications where highly targeted shapes are desired. We propose an architecture which simultaneously predicts the intermediate shape lying in the nonlinear transformation pathway between the undeformed and desired flow shape, then learns the causal action--the single pillar which results in the deformation of the flow--one at a time. The learning of stage-wise transformations provides deep insights into the physical flow deformation. Results show that under the current framework, our model is able to predict a sequence of pillars that reconstructs the flow shape which highly resembles the desired shape.

Adversarially Robust Generalization Requires More Data

Machine learning models are often susceptible to adversarial perturbations of their inputs. Even small perturbations can cause state-of-the-art classifiers with high "standard" accuracy to produce an incorrect prediction with high confidence. To better understand this phenomenon, we study adversarially robust learning from the viewpoint of generalization. We show that already in a simple natural data model, the sample complexity of robust learning can be significantly larger than that of "standard" learning. This gap is information theoretic and holds irrespective of the training algorithm or the model family. We complement our theoretical results with experiments on popular image classification datasets and show that a similar gap exists here as well. We postulate that the difficulty of training robust classifiers stems, at least partially, from this inherently larger sample complexity.

A Hybrid Solution to improve Iteration Efficiency in the Distributed Learning

Currently, many machine learning algorithms contain lots of iterations. When it comes to existing large-scale distributed systems, some slave nodes may break down or have lower efficiency. Therefore traditional machine learning algorithm may fail because of the instability of distributed system.We presents a hybrid approach which not only own a high fault-tolerant but also achieve a balance of performance and efficiency.For each iteration, the result of slow machines will be abandoned. Then, we discuss the relationship between accuracy and abandon rate. Next we debate the convergence speed of this process. Finally, our experiments demonstrate our idea can dramatically reduce calculation time and be used in many platforms.

A complex network approach to stylometry

Statistical methods have been widely employed to study the fundamental properties of language. In recent years, methods from complex and dynamical systems proved useful to create several language models. Despite the large amount of studies devoted to represent texts with physical models, only a limited number of studies have shown how the properties of the underlying physical systems can be employed to improve the performance of natural language processing tasks. In this paper, I address this problem by devising complex networks methods that are able to improve the performance of current statistical methods. Using a fuzzy classification strategy, I show that the topological properties extracted from texts complement the traditional textual description. In several cases, the performance obtained with hybrid approaches outperformed the results obtained when only traditional or networked methods were used. Because the proposed model is generic, the framework devised here could be straightforwardly used to study similar textual applications where the topology plays a pivotal role in the description of the interacting agents.

Automated Generation of Multilingual Clusters for the Evaluation of Distributed Representations

We propose a language-agnostic way of automatically generating sets of semantically similar clusters of entities along with sets of "outlier" elements, which may then be used to perform an intrinsic evaluation of word embeddings in the outlier detection task. We used our methodology to create a gold-standard dataset, which we call WikiSem500, and evaluated multiple state-of-the-art embeddings. The results show a correlation between performance on this dataset and performance on sentiment analysis.

Importance Weighting and Variational Inference

Recent work used importance sampling ideas for better variational bounds on likelihoods. We clarify the applicability of these ideas to pure probabilistic inference, by showing the resulting Importance Weighted Variational Inference (IWVI) technique is an instance of augmented variational inference, thus identifying the looseness in previous work. Experiments confirm IWVI's practicality for probabilistic inference. As a second contribution, we investigate inference with elliptical distributions, which improves accuracy in low dimensions, and convergence in high dimensions.

Safe and Efficient Off-Policy Reinforcement Learning

In this work, we take a fresh look at some old and new algorithms for off-policy, return-based reinforcement learning. Expressing these in a common form, we derive a novel algorithm, Retrace(lambda), with three desired properties: (1) it has low variance; (2) it safely uses samples collected from any behaviour policy, whatever its degree of "off-policyness"; and (3) it is efficient as it makes the best use of samples collected from near on-policy behaviour policies. We analyse the contractive nature of the related operator under both off-policy policy evaluation and control settings and derive online sample-based algorithms. We believe this is the first return-based off-policy control algorithm converging a.s. to Q* without the GLIE assumption (Greedy in the Limit with Infinite Exploration). As a corollary, we prove the convergence of Watkins' Q(lambda), which was an open problem since 1989. We illustrate the benefits of Retrace(lambda) on a standard suite of Atari 2600 games.

Diversified Top-k Partial MaxSAT Solving

We introduce a diversified top-k partial MaxSAT problem, a combination of partial MaxSAT problem and enumeration problem. Given a partial MaxSAT formula F and a positive integer k, the diversified top-k partial MaxSAT is to find k maximal solutions for F such that the k maximal solutions satisfy the maximum number of soft clauses of F. This problem can be widely used in many applications including community detection, sensor place, motif discovery, and combinatorial testing. We prove the problem is NP-hard and propose an approach for solving the problem. The concrete idea of the approach is to design an encoding EE which reduces diversified top-k partial MaxSAT problem into partial MaxSAT problem, and then solve the resulting problem with state-of-art solvers. In addition, we present an algorithm MEMKC exactly solving the diversified top-k partial MaxSAT. Through several experiments we show that our approach can be successfully applied to the interesting problem.

Generating Images from Captions with Attention

Motivated by the recent progress in generative models, we introduce a model that generates images from natural language descriptions. The proposed model iteratively draws patches on a canvas, while attending to the relevant words in the description. After training on Microsoft COCO, we compare our model with several baseline generative models on image generation and retrieval tasks. We demonstrate that our model produces higher quality samples than other approaches and generates images with novel scene compositions corresponding to previously unseen captions in the dataset.

Online Learning with Switching Costs and Other Adaptive Adversaries

We study the power of different types of adaptive (nonoblivious) adversaries in the setting of prediction with expert advice, under both full information and bandit feedback. We measure the player's performance using a new notion of regret, also known as policy regret, which better captures the adversary's adaptiveness to the player's behavior. In a setting where losses are allowed to drift, we characterize ---in a nearly complete manner--- the power of adaptive adversaries with bounded memories and switching costs. In particular, we show that with switching costs, the attainable rate with bandit feedback is $\Theta(T^{2/3})$. Interestingly, this rate is significantly worse than the $\Theta(\sqrt{T})$ rate attainable with switching costs in the full information case. Via a novel reduction from experts to bandits, we also show that a bounded memory adversary can force $\Theta(T^{2/3})$ regret even in the full information case, proving that switching costs are easier to control than bounded memory adversaries. Our lower bounds rely on a new stochastic adversary strategy that generates loss processes with strong dependencies.

Measuring Thematic Fit with Distributional Feature Overlap

In this paper, we introduce a new distributional method for modeling predicate-argument thematic fit judgments. We use a syntax-based DSM to build a prototypical representation of verb-specific roles: for every verb, we extract the most salient second order contexts for each of its roles (i.e. the most salient dimensions of typical role fillers), and then we compute thematic fit as a weighted overlap between the top features of candidate fillers and role prototypes. Our experiments show that our method consistently outperforms a baseline re-implementing a state-of-the-art system, and achieves better or comparable results to those reported in the literature for the other unsupervised systems. Moreover, it provides an explicit representation of the features characterizing verb-specific semantic roles.

Efficiently Discovering Hammock Paths from Induced Similarity Networks

Similarity networks are important abstractions in many information management applications such as recommender systems, corpora analysis, and medical informatics. For instance, by inducing similarity networks between movies rated similarly by users, or between documents containing common terms, and or between clinical trials involving the same themes, we can aim to find the global structure of connectivities underlying the data, and use the network as a basis to make connections between seemingly disparate entities. In the above applications, composing similarities between objects of interest finds uses in serendipitous recommendation, in storytelling, and in clinical diagnosis, respectively. We present an algorithmic framework for traversing similarity paths using the notion of `hammock' paths which are generalization of traditional paths. Our framework is exploratory in nature so that, given starting and ending objects of interest, it explores candidate objects for path following, and heuristics to admissibly estimate the potential for paths to lead to a desired destination. We present three diverse applications: exploring movie similarities in the Netflix dataset, exploring abstract similarities across the PubMed corpus, and exploring description similarities in a database of clinical trials. Experimental results demonstrate the potential of our approach for unstructured knowledge discovery in similarity networks.

Epsilon-Best-Arm Identification in Pay-Per-Reward Multi-Armed Bandits

We study epsilon-best-arm identification, in a setting where during the exploration phase, the cost of each arm pull is proportional to the expected future reward of that arm. We term this setting Pay-Per-Reward. We provide an algorithm for this setting, that with a high probability returns an epsilon-best arm, while incurring a cost that depends only linearly on the total expected reward of all arms, and does not depend at all on the number of arms. Under mild assumptions, the algorithm can be applied also to problems with infinitely many arms.

Invertibility of Convolutional Generative Networks from Partial Measurements

In this work, we present new theoretical results on convolutional generative neural networks, in particular their invertibility (i.e., the recovery of input latent code given the network output). The study of network inversion problem is motivated by image inpainting and the mode collapse problem in training GAN. Network inversion is highly non-convex, and thus is typically computationally intractable and without optimality guarantees. However, we rigorously prove that, under some mild technical assumptions, the input of a two-layer convolutional generative network can be deduced from the network output efficiently using simple gradient descent. This new theoretical finding implies that the mapping from the low- dimensional latent space to the high-dimensional image space is bijective (i.e., one-to-one). In addition, the same conclusion holds even when the network output is only partially observed (i.e., with missing pixels). Our theorems hold for 2-layer convolutional generative network with ReLU as the activation function, but we demonstrate empirically that the same conclusion extends to multi-layer networks and networks with other activation functions, including the leaky ReLU, sigmoid and tanh.

Linguistic Descriptions for Automatic Generation of Textual Short-Term Weather Forecasts on Real Prediction Data

We present in this paper an application which automatically generates textual short-term weather forecasts for every municipality in Galicia (NW Spain), using the real data provided by the Galician Meteorology Agency (MeteoGalicia). This solution combines in an innovative way computing with perceptions techniques and strategies for linguistic description of data together with a natural language generation (NLG) system. The application, named GALiWeather, extracts relevant information from weather forecast input data and encodes it into intermediate descriptions using linguistic variables and temporal references. These descriptions are later translated into natural language texts by the natural language generation system. The obtained forecast results have been thoroughly validated by an expert meteorologist from MeteoGalicia using a quality assessment methodology which covers two key dimensions of a text: the accuracy of its content and the correctness of its form. Following this validation GALiWeather will be released as a real service offering custom forecasts for a wide public.

Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling

We study the problem of 3D object generation. We propose a novel framework, namely 3D Generative Adversarial Network (3D-GAN), which generates 3D objects from a probabilistic space by leveraging recent advances in volumetric convolutional networks and generative adversarial nets. The benefits of our model are three-fold: first, the use of an adversarial criterion, instead of traditional heuristic criteria, enables the generator to capture object structure implicitly and to synthesize high-quality 3D objects; second, the generator establishes a mapping from a low-dimensional probabilistic space to the space of 3D objects, so that we can sample objects without a reference image or CAD models, and explore the 3D object manifold; third, the adversarial discriminator provides a powerful 3D shape descriptor which, learned without supervision, has wide applications in 3D object recognition. Experiments demonstrate that our method generates high-quality 3D objects, and our unsupervisedly learned features achieve impressive performance on 3D object recognition, comparable with those of supervised learning methods.

On the Interpretability of Conditional Probability Estimates in the Agnostic Setting

Many classification algorithms produce confidence measures in the form of conditional probability of labels given the features of the target instance. It is desirable to be make these confidence measures calibrated or consistent, in the sense that they correctly capture the belief of the algorithm in the label output. For instance, if the algorithm outputs a label with confidence measure $p$ for $n$ times, then the output label should be correct approximately $np$ times overall. Calibrated confidence measures lead to higher interpretability by humans and computers and enable downstream analysis or processing. In this paper, we formally characterize the consistency of confidence measures and prove a PAC-style uniform convergence result for the consistency of confidence measures. We show that finite VC-dimension is sufficient for guaranteeing the consistency of confidence measures produced by empirically consistent classifiers. Our result also implies that we can calibrate confidence measures produced by any existing algorithms with monotonic functions, and still get the same generalization guarantee on consistency.

Deep AutoRegressive Networks

We introduce a multilayer deep generative model capable of learning hierarchies of sparse distributed representations from data. The model consists of several layers of stochastic units, with autoregressive connections within each layer, which allows for efficient exact sampling. We train the model efficiently using an algorithm derived from the Minimum Description Length (MDL) principle, which minimizes the amount of information contained in the joint vector of data and hidden unit configurations for the training set. As we are not given the hidden unit configurations corresponding to the training data, we use a feedforward network to map data vectors to configurations of hidden units that are jointly probable with them and train it jointly with the model. Our approach can also be seen as maximizing a lower bound on the log-likelihood, with the feedforward network implementing approximate inference.

Turing Test for the Internet of Things

How smart is your kettle? How smart are things in your kitchen, your house, your neighborhood, on the internet? With the advent of Internet of Things, and the move of making devices `smart' by utilizing AI, a natural question arrises, how can we evaluate the progress. The standard way of evaluating AI is through the Turing Test. While Turing Test was designed for AI; the device that it was tailored to was a computer. Applying the test to variety of devices that constitute Internet of Things poses a number of challenges which could be addressed through a number of adaptations.

A Fully Convolutional Neural Network for Speech Enhancement

In hearing aids, the presence of babble noise degrades hearing intelligibility of human speech greatly. However, removing the babble without creating artifacts in human speech is a challenging task in a low SNR environment. Here, we sought to solve the problem by finding a `mapping' between noisy speech spectra and clean speech spectra via supervised learning. Specifically, we propose using fully Convolutional Neural Networks, which consist of lesser number of parameters than fully connected networks. The proposed network, Redundant Convolutional Encoder Decoder (R-CED), demonstrates that a convolutional network can be 12 times smaller than a recurrent network and yet achieves better performance, which shows its applicability for an embedded system: the hearing aids.

Bandits meet Computer Architecture: Designing a Smartly-allocated Cache

In many embedded systems, such as imaging sys- tems, the system has a single designated purpose, and same threads are executed repeatedly. Profiling thread behavior, allows the system to allocate each thread its resources in a way that improves overall system performance. We study an online resource al- locationproblem,wherearesourcemanagersimulta- neously allocates resources (exploration), learns the impact on the different consumers (learning) and im- proves allocation towards optimal performance (ex- ploitation). We build on the rich framework of multi- armed bandits and present online and offline algo- rithms. Through extensive experiments with both synthetic data and real-world cache allocation to threads we show the merits and properties of our al- gorithms

Budget Constraints in Prediction Markets

We give a detailed characterization of optimal trades under budget constraints in a prediction market with a cost-function-based automated market maker. We study how the budget constraints of individual traders affect their ability to impact the market price. As a concrete application of our characterization, we give sufficient conditions for a property we call budget additivity: two traders with budgets B and B' and the same beliefs would have a combined impact equal to a single trader with budget B+B'. That way, even if a single trader cannot move the market much, a crowd of like-minded traders can have the same desired effect. When the set of payoff vectors associated with outcomes, with coordinates corresponding to securities, is affinely independent, we obtain that a generalization of the heavily-used logarithmic market scoring rule is budget additive, but the quadratic market scoring rule is not. Our results may be used both descriptively, to understand if a particular market maker is affected by budget constraints or not, and prescriptively, as a recipe to construct markets.

A Unified Approach for Learning the Parameters of Sum-Product Networks

We present a unified approach for learning the parameters of Sum-Product networks (SPNs). We prove that any complete and decomposable SPN is equivalent to a mixture of trees where each tree corresponds to a product of univariate distributions. Based on the mixture model perspective, we characterize the objective function when learning SPNs based on the maximum likelihood estimation (MLE) principle and show that the optimization problem can be formulated as a signomial program. Both the projected gradient descent (PGD) and the exponentiated gradient (EG) in this setting can be viewed as first order approximations of the signomial program after proper transformation of the objective function. Based on the signomial program formulation, we construct two parameter learning algorithms for SPNs by using sequential monomial approximations (SMA) and the concave-convex procedure (CCCP), respectively. The two proposed methods naturally admit multiplicative updates, hence effectively avoiding the projection operation. With the help of the a unified framework, we also show an intrinsic connection between CCCP and Expectation Maximization (EM), where EM turns out to be another relaxation of the signomial program. Extensive experiments on 20 data sets demonstrate the effectiveness and efficiency of the two proposed approaches for learning SPNs. We also show that the proposed methods can improve the performance of structure learning and yield state-of-the-art results.

Gaussian Process Prior Variational Autoencoders

Variational autoencoders (VAE) are a powerful and widely-used class of models to learn complex data distributions in an unsupervised fashion. One important limitation of VAEs is the prior assumption that latent sample representations are independent and identically distributed. However, for many important datasets, such as time-series of images, this assumption is too strong: accounting for covariances between samples, such as those in time, can yield to a more appropriate model specification and improve performance in downstream tasks. In this work, we introduce a new model, the Gaussian Process (GP) Prior Variational Autoencoder (GPPVAE), to specifically address this issue. The GPPVAE aims to combine the power of VAEs with the ability to model correlations afforded by GP priors. To achieve efficient inference in this new class of models, we leverage structure in the covariance matrix, and introduce a new stochastic backpropagation strategy that allows for computing stochastic gradients in a distributed and low-memory fashion. We show that our method outperforms conditional VAEs (CVAEs) and an adaptation of standard VAEs in two image data applications.

Structured Reachability Analysis for Markov Decision Processes

Recent research in decision theoretic planning has focussed on making the solution of Markov decision processes (MDPs) more feasible. We develop a family of algorithms for structured reachability analysis of MDPs that are suitable when an initial state (or set of states) is known. Using compact, structured representations of MDPs (e.g., Bayesian networks), our methods, which vary in the tradeoff between complexity and accuracy, produce structured descriptions of (estimated) reachable states that can be used to eliminate variables or variable values from the problem description, reducing the size of the MDP and making it easier to solve. One contribution of our work is the extension of ideas from GRAPHPLAN to deal with the distributed nature of action representations typically embodied within Bayes nets and the problem of correlated action effects. We also demonstrate that our algorithm can be made more complete by using k-ary constraints instead of binary constraints. Another contribution is the illustration of how the compact representation of reachability constraints can be exploited by several existing (exact and approximate) abstraction algorithms for MDPs.

Spectral Filtering for General Linear Dynamical Systems

We give a polynomial-time algorithm for learning latent-state linear dynamical systems without system identification, and without assumptions on the spectral radius of the system's transition matrix. The algorithm extends the recently introduced technique of spectral filtering, previously applied only to systems with a symmetric transition matrix, using a novel convex relaxation to allow for the efficient identification of phases.

Storytelling Agents with Personality and Adaptivity

We explore the expression of personality and adaptivity through the gestures of virtual agents in a storytelling task. We conduct two experiments using four different dialogic stories. We manipulate agent personality on the extraversion scale, whether the agents adapt to one another in their gestural performance and agent gender. Our results show that subjects are able to perceive the intended variation in extraversion between different virtual agents, independently of the story they are telling and the gender of the agent. A second study shows that subjects also prefer adaptive to nonadaptive virtual agents.

Online Sequence Training of Recurrent Neural Networks with Connectionist Temporal Classification

Connectionist temporal classification (CTC) based supervised sequence training of recurrent neural networks (RNNs) has shown great success in many machine learning areas including end-to-end speech and handwritten character recognition. For the CTC training, however, it is required to unroll the RNN by the length of an input sequence. This unrolling requires a lot of memory and hinders a small footprint implementation of online learning or adaptation. Furthermore, the length of training sequences is usually not uniform, which makes parallel training with multiple sequences inefficient on shared memory models such as graphics processing units (GPUs). In this work, we introduce an expectation-maximization (EM) based online CTC algorithm that enables unidirectional RNNs to learn sequences that are longer than the amount of unrolling. The RNNs can also be trained to process an infinitely long input sequence without pre-segmentation or external reset. Moreover, the proposed approach allows efficient parallel training on GPUs. For evaluation, end-to-end speech recognition examples are presented on the Wall Street Journal (WSJ) corpus.