Datasets:

chreh

/

peer_read_neurips

like 0

Distributed Transfer Linear Support Vector Machines

Transfer learning has been developed to improve the performances of different but related tasks in machine learning. However, such processes become less efficient with the increase of the size of training data and the number of tasks. Moreover, privacy can be violated as some tasks may contain sensitive and private data, which are communicated between nodes and tasks. We propose a consensus-based distributed transfer learning framework, where several tasks aim to find the best linear support vector machine (SVM) classifiers in a distributed network. With alternating direction method of multipliers, tasks can achieve better classification accuracies more efficiently and privately, as each node and each task train with their own data, and only decision variables are transferred between different tasks and nodes. Numerical experiments on MNIST datasets show that the knowledge transferred from the source tasks can be used to decrease the risks of the target tasks that lack training data or have unbalanced training labels. We show that the risks of the target tasks in the nodes without the data of the source tasks can also be reduced using the information transferred from the nodes who contain the data of the source tasks. We also show that the target tasks can enter and leave in real-time without rerunning the whole algorithm.

RAPID: A Reachable Anytime Planner for Imprecisely-sensed Domains

Despite the intractability of generic optimal partially observable Markov decision process planning, there exist important problems that have highly structured models. Previous researchers have used this insight to construct more efficient algorithms for factored domains, and for domains with topological structure in the flat state dynamics model. In our work, motivated by findings from the education community relevant to automated tutoring, we consider problems that exhibit a form of topological structure in the factored dynamics model. Our Reachable Anytime Planner for Imprecisely-sensed Domains (RAPID) leverages this structure to efficiently compute a good initial envelope of reachable states under the optimal MDP policy in time linear in the number of state variables. RAPID performs partially-observable planning over the limited envelope of states, and slowly expands the state space considered as time allows. RAPID performs well on a large tutoring-inspired problem simulation with 122 state variables, corresponding to a flat state space of over 10^30 states.

Value-Directed Belief State Approximation for POMDPs

We consider the problem belief-state monitoring for the purposes of implementing a policy for a partially-observable Markov decision process (POMDP), specifically how one might approximate the belief state. Other schemes for belief-state approximation (e.g., based on minimixing a measures such as KL-diveregence between the true and estimated state) are not necessarily appropriate for POMDPs. Instead we propose a framework for analyzing value-directed approximation schemes, where approximation quality is determined by the expected error in utility rather than by the error in the belief state itself. We propose heuristic methods for finding good projection schemes for belief state estimation - exhibiting anytime characteristics - given a POMDP value fucntion. We also describe several algorithms for constructing bounds on the error in decision quality (expected utility) associated with acting in accordance with a given belief state approximation.

To Fall Or Not To Fall: A Visual Approach to Physical Stability Prediction

Understanding physical phenomena is a key competence that enables humans and animals to act and interact under uncertain perception in previously unseen environments containing novel object and their configurations. Developmental psychology has shown that such skills are acquired by infants from observations at a very early stage.

Spectral Norm Regularization of Orthonormal Representations for Graph Transduction

Recent literature~\cite{ando} suggests that embedding a graph on an unit sphere leads to better generalization for graph transduction. However, the choice of optimal embedding and an efficient algorithm to compute the same remains open. In this paper, we show that orthonormal representations, a class of unit-sphere graph embeddings are PAC learnable. Existing PAC-based analysis do not apply as the VC dimension of the function class is infinite. We propose an alternative PAC-based bound, which do not depend on the VC dimension of the underlying function class, but is related to the famous Lov\'{a}sz~$\vartheta$ function. The main contribution of the paper is SPORE, a SPectral regularized ORthonormal Embedding for graph transduction, derived from the PAC bound. SPORE is posed as a non-smooth convex function over an \emph{elliptope}. These problems are usually solved as semi-definite programs (SDPs) with time complexity $O(n^6)$. We present, Infeasible Inexact proximal~(IIP): an Inexact proximal method which performs subgradient procedure on an approximate projection, not necessarily feasible. IIP is more scalable than SDP, has an $O(\frac{1}{\sqrt{T}})$ convergence, and is generally applicable whenever a suitable approximate projection is available. We use IIP to compute SPORE where the approximate projection step is computed by FISTA, an accelerated gradient descent procedure. We show that the method has a convergence rate of $O(\frac{1}{\sqrt{T}})$. The proposed algorithm easily scales to 1000's of vertices, while the standard SDP computation does not scale beyond few hundred vertices. Furthermore, the analysis presented here easily extends to the multiple graph setting.

Neural Universal Discrete Denoiser

We present a new framework of applying deep neural networks (DNN) to devise a universal discrete denoiser. Unlike other approaches that utilize supervised learning for denoising, we do not require any additional training data. In such setting, while the ground-truth label, i.e., the clean data, is not available, we devise ``pseudo-labels'' and a novel objective function such that DNN can be trained in a same way as supervised learning to become a discrete denoiser. We experimentally show that our resulting algorithm, dubbed as Neural DUDE, significantly outperforms the previous state-of-the-art in several applications with a systematic rule of choosing the hyperparameter, which is an attractive feature in practice.

Tight Bounds for Influence in Diffusion Networks and Application to Bond Percolation and Epidemiology

In this paper, we derive theoretical bounds for the long-term influence of a node in an Independent Cascade Model (ICM). We relate these bounds to the spectral radius of a particular matrix and show that the behavior is sub-critical when this spectral radius is lower than 1. More specifically, we point out that, in general networks, the sub-critical regime behaves in O(sqrt(n)) where n is the size of the network, and that this upper bound is met for star-shaped networks. We apply our results to epidemiology and percolation on arbitrary networks, and derive a bound for the critical value beyond which a giant connected component arises. Finally, we show empirically the tightness of our bounds for a large family of networks.

Controlling privacy in recommender systems

Recommender systems involve an inherent trade-off between accuracy of recommendations and the extent to which users are willing to release information about their preferences. In this paper, we explore a two-tiered notion of privacy where there is a small set of <code>public'' users who are willing to share their preferences openly, and a large set of</code>private'' users who require privacy guarantees. We show theoretically and demonstrate empirically that a moderate number of public users with no access to private user information already suffices for reasonable accuracy. Moreover, we introduce a new privacy concept for gleaning relational information from private users while maintaining a first order deniability. We demonstrate gains from controlled access to private user preferences.

A flexible model for training action localization with varying levels of supervision

Spatio-temporal action detection in videos is typically addressed in a fully-supervised setup with manual annotation of training videos required at every frame. Since such annotation is extremely tedious and prohibits scalability, there is a clear need to minimize the amount of manual supervision. In this work we propose a unifying framework that can handle and combine varying types of less demanding weak supervision. Our model is based on discriminative clustering and integrates different types of supervision as constraints on the optimization. We investigate applications of such a model to training setups with alternative supervisory signals ranging from video-level class labels over temporal points or sparse action bounding boxes to the full per-frame annotation of action bounding boxes. Experiments on the challenging UCF101-24 and DALY datasets demonstrate competitive performance of our method at a fraction of supervision used by previous methods. The flexibility of our model enables joint learning from data with different levels of annotation. Experimental results demonstrate a significant gain by adding a few fully supervised examples to otherwise weakly labeled videos.

Simultaneous Low-rank Component and Graph Estimation for High-dimensional Graph Signals: Application to Brain Imaging

We propose an algorithm to uncover the intrinsic low-rank component of a high-dimensional, graph-smooth and grossly-corrupted dataset, under the situations that the underlying graph is unknown. Based on a model with a low-rank component plus a sparse perturbation, and an initial graph estimation, our proposed algorithm simultaneously learns the low-rank component and refines the graph. Our evaluations using synthetic and real brain imaging data in unsupervised and supervised classification tasks demonstrate encouraging performance.

GPU Activity Prediction using Representation Learning

GPU activity prediction is an important and complex problem. This is due to the high level of contention among thousands of parallel threads. This problem was mostly addressed using heuristics. We propose a representation learning approach to address this problem. We model any performance metric as a temporal function of the executed instructions with the intuition that the flow of instructions can be identified as distinct activities of the code. Our experiments show high accuracy and non-trivial predictive power of representation learning on a benchmark.

Sparse PCA via Covariance Thresholding

In sparse principal component analysis we are given noisy observations of a low-rank matrix of dimension $n\times p$ and seek to reconstruct it under additional sparsity assumptions. In particular, we assume here that the principal components $\bv_1,\dots,\bv_r$ have at most $k_1, \cdots, k_q$ non-zero entries respectively, and study the high-dimensional regime in which $p$ is of the same order as $n$. In an influential paper, Johnstone and Lu \cite{johnstone2004sparse} introduced a simple algorithm that estimates the support of the principal vectors $\bv_1,\dots,\bv_r$ by the largest entries in the diagonal of the empirical covariance. This method can be shown to succeed with high probability if $k_q \le C_1\sqrt{n/\log p}$, and to fail with high probability if $k_q\ge C_2 \sqrt{n/\log p}$ for two constants $0 < C_1,C_2 < \infty$. Despite a considerable amount of work over the last ten years, no practical algorithm exists with provably better support recovery guarantees. Here we analyze a covariance thresholding algorithm that was recently proposed by Krauthgamer, Nadler and Vilenchik \cite{KrauthgamerSPCA}. We confirm empirical evidence presented by these authors and rigorously prove that the algorithm succeeds with high probability for $k$ of order $\sqrt{n}$. Recent conditional lower bounds \cite{berthet2013computational} suggest that it might be impossible to do significantly better. The key technical component of our analysis develops new bounds on the norm of kernel random matrices, in regimes that were not considered before.

Web Similarity

Normalized web distance (NWD) is a similarity or normalized semantic distance based on the World Wide Web or any other large electronic database, for instance Wikipedia, and a search engine that returns reliable aggregate page counts. For sets of search terms the NWD gives a similarity on a scale from 0 (identical) to 1 (completely different). The NWD approximates the similarity according to all (upper semi)computable properties. We develop the theory and give applications. The derivation of the NWD method is based on Kolmogorov complexity.

Variational Dropout Sparsifies Deep Neural Networks

We explore recently proposed variational dropout technique which provided an elegant Bayesian interpretation to dropout. We extend variational dropout to the case when dropout rate is unknown and show that it can be found by optimizing evidence variational lower bound. We show that it is possible to assign and find individual dropout rates to each connection in DNN. Interestingly such assignment leads to extremely sparse solutions both in fully-connected and convolutional layers. This effect is similar to automatic relevance determination (ARD) effect in empirical Bayes but has a number of advantages. We report up to 128 fold compression of popular architectures without a large loss of accuracy providing additional evidence to the fact that modern deep architectures are very redundant.

Homotopy based algorithms for $\ell_0$-regularized least-squares

Sparse signal approximation can be formulated as the mixed $\ell_2$-$\ell_0$ minimization problem $\min_x J(x;\lambda)=\|y-Ax\|_2^2+\lambda\|x\|_0$. We propose two heuristic search algorithms to minimize J for a continuum of $\lambda$-values, yielding a sequence of coarse to fine approximations. Continuation Single Best Replacement is a bidirectional greedy algorithm adapted from the Single Best Replacement algorithm previously proposed for minimizing J for fixed $\lambda$. $\ell_0$ regularization path track is a more complex algorithm exploiting that the $\ell_2$-$\ell_0$ regularization path is piecewise constant with respect to $\lambda$. Tracking the $\ell_0$ regularization path is done in a sub-optimal manner by maintaining (i) a list of subsets that are candidates to be solution supports for decreasing $\lambda$'s and (ii) the list of critical $\lambda$-values around which the solution changes. Both algorithms gradually construct the $\ell_0$ regularization path by performing single replacements, i.e., adding or removing a dictionary atom from a subset. A straightforward adaptation of these algorithms yields sub-optimal solutions to $\min_x \|y-Ax\|_2^2$ subject to $\|x\|_0\leq k$ for contiguous values of $k\geq 0$ and to $\min_x \|x\|_0$ subject to $\|y-Ax\|_2^2\leq\varepsilon$ for continuous values of $\varepsilon$. Numerical simulations show the effectiveness of the algorithms on a difficult sparse deconvolution problem inducing a highly correlated dictionary A.

Towards the Limit of Network Quantization

Network quantization is one of network compression techniques employed to reduce the redundancy of deep neural networks. It compresses the size of the storage for a large number of network parameters in a neural network by quantizing them and encoding the quantized values into binary codewords of smaller sizes. In this paper, we aim to design network quantization schemes that minimize the expected loss due to quantization while maximizing the compression ratio. To this end, we analyze the quantitative relation of quantization errors to the loss function of a neural network and identify that the Hessian-weighted distortion measure is locally the right objective function that we need to optimize for minimizing the loss due to quantization. As a result, Hessian-weighted k-means clustering is proposed for clustering network parameters to quantize when fixed-length binary encoding follows. When optimal variable-length binary codes, e.g., Huffman codes, are employed for further compression of quantized values after clustering, we derive that the network quantization problem can be related to the entropy-constrained scalar quantization (ECSQ) problem in information theory and consequently propose two solutions of ECSQ for network quantization, i.e., uniform quantization and an iterative algorithm similar to Lloyd's algorithm for k-means clustering. Finally, using the simple uniform quantization followed by Huffman coding, our experiment results show that the compression ratios of 51.25, 22.17 and 40.65 are achievable (i.e., the sizes of the compressed models are 1.95%, 4.51% and 2.46% of the original model sizes) for LeNet, ResNet and AlexNet, respectively, at no or marginal performance loss.

Bilevel learning of the Group Lasso structure

Regression with group-sparsity penalty plays a central role in high-dimensional prediction problems. Most of existing methods require the group structure to be known a priori. In practice, this may be a too strong assumption, potentially hampering the effectiveness of the regularization method. To circumvent this issue, we present a method to estimate the group structure by means of a continuous bilevel optimization problem where the data is split into training and validation sets. Our approach relies on an approximation scheme where the lower level problem is replaced by a smooth dual forward-backward algorithm with Bregman distances. We provide guarantees regarding the convergence of the approximate procedure to the exact problem and demonstrate the well behaviour of the proposed method on synthetic experiments. Finally, a preliminary application to genes expression data is tackled with the purpose of unveiling functional groups.

Partial Hard Thresholding: Towards A Principled Analysis of Support Recovery

In machine learning and compressed sensing, it is of central importance to understand when a tractable algorithm recovers the support of a sparse signal from its compressed measurements. In this paper, we present a principled analysis on the support recovery performance for a family of hard thresholding algorithms. To this end, we appeal to the partial hard thresholding (PHT) operator proposed recently by Jain et al. [IEEE Trans. Information Theory, 2017]. We show that under proper conditions, PHT recovers an arbitrary "s"-sparse signal within O(s kappa log kappa) iterations where "kappa" is an appropriate condition number. Specifying the PHT operator, we obtain the best known result for hard thresholding pursuit and orthogonal matching pursuit with replacement. Experiments on the simulated data complement our theoretical findings and also illustrate the effectiveness of PHT compared to other popular recovery methods.

Towards balanced clustering - part 1 (preliminaries)

The article contains a preliminary glance at balanced clustering problems. Basic balanced structures and combinatorial balanced problems are briefly described. A special attention is targeted to various balance/unbalance indices (including some new versions of the indices): by cluster cardinality, by cluster weights, by inter-cluster edge/arc weights, by cluster element structure (for element multi-type clustering). Further, versions of optimization clustering problems are suggested (including multicriteria problem formulations). Illustrative numerical examples describe calculation of balance indices and element multi-type balance clustering problems (including example for design of student teams).

A Review of Verbal and Non-Verbal Human-Robot Interactive Communication

In this paper, an overview of human-robot interactive communication is presented, covering verbal as well as non-verbal aspects of human-robot interaction. Following a historical introduction, and motivation towards fluid human-robot communication, ten desiderata are proposed, which provide an organizational axis both of recent as well as of future research on human-robot communication. Then, the ten desiderata are examined in detail, culminating to a unifying discussion, and a forward-looking conclusion.

Local SGD with Periodic Averaging: Tighter Analysis and Adaptive Synchronization

Communication overhead is one of the key challenges that hinders the scalability of distributed optimization algorithms. In this paper, we study local distributed SGD, where data is partitioned among computation nodes, and the computation nodes perform local updates with periodically exchanging the model among the workers to perform averaging. While local SGD is empirically shown to provide promising results, a theoretical understanding of its performance remains open. In this paper, we strengthen convergence analysis for local SGD, and show that local SGD can be far less expensive and applied far more generally than current theory suggests. Specifically, we show that for loss functions that satisfy the Polyak-Kojasiewicz condition, $O((pT)^{1/3})$ rounds of communication suffice to achieve a linear speed up, that is, an error of $O(1/pT)$, where $T$ is the total number of model updates at each worker. This is in contrast with previous work which required higher number of communication rounds, as well as was limited to strongly convex loss functions, for a similar asymptotic performance. We also develop an adaptive synchronization scheme that provides a general condition for linear speed up. Finally, we validate the theory with experimental results, running over AWS EC2 clouds and an internal GPUs cluster.

A Semidefinite Programming Based Search Strategy for Feature Selection with Mutual Information Measure

Feature subset selection, as a special case of the general subset selection problem, has been the topic of a considerable number of studies due to the growing importance of data-mining applications. In the feature subset selection problem there are two main issues that need to be addressed: (i) Finding an appropriate measure function than can be fairly fast and robustly computed for high-dimensional data. (ii) A search strategy to optimize the measure over the subset space in a reasonable amount of time. In this article mutual information between features and class labels is considered to be the measure function. Two series expansions for mutual information are proposed, and it is shown that most heuristic criteria suggested in the literature are truncated approximations of these expansions. It is well-known that searching the whole subset space is an NP-hard problem. Here, instead of the conventional sequential search algorithms, we suggest a parallel search strategy based on semidefinite programming (SDP) that can search through the subset space in polynomial time. By exploiting the similarities between the proposed algorithm and an instance of the maximum-cut problem in graph theory, the approximation ratio of this algorithm is derived and is compared with the approximation ratio of the backward elimination method. The experiments show that it can be misleading to judge the quality of a measure solely based on the classification accuracy, without taking the effect of the non-optimum search strategy into account.

Review Networks for Caption Generation

We propose a novel module, the reviewer module, to improve the encoder-decoder learning framework. The reviewer module is generic, and can be plugged into an existing encoder-decoder model. The reviewer module performs a number of review steps with attention mechanism on the encoder hidden states, and outputs a fact vector after each review step; the fact vectors are used as the input of the attention mechanism in the decoder. We show that the conventional encoder-decoders are a special case of our framework. Empirically, we show that our framework can improve over state-of-the-art encoder-decoder systems on the tasks of image captioning and source code captioning.

Learning Longer Memory in Recurrent Neural Networks

Recurrent neural network is a powerful model that learns temporal patterns in sequential data. For a long time, it was believed that recurrent networks are difficult to train using simple optimizers, such as stochastic gradient descent, due to the so-called vanishing gradient problem. In this paper, we show that learning longer term patterns in real data, such as in natural language, is perfectly possible using gradient descent. This is achieved by using a slight structural modification of the simple recurrent neural network architecture. We encourage some of the hidden units to change their state slowly by making part of the recurrent weight matrix close to identity, thus forming kind of a longer term memory. We evaluate our model in language modeling experiments, where we obtain similar performance to the much more complex Long Short Term Memory (LSTM) networks (Hochreiter &amp; Schmidhuber, 1997).

Parallel Multi-Dimensional LSTM, With Application to Fast Biomedical Volumetric Image Segmentation

Convolutional Neural Networks (CNNs) can be shifted across 2D images or 3D videos to segment them. They have a fixed input size and typically perceive only small local contexts of the pixels to be classified as foreground or background. In contrast, Multi-Dimensional Recurrent NNs (MD-RNNs) can perceive the entire spatio-temporal context of each pixel in a few sweeps through all pixels, especially when the RNN is a Long Short-Term Memory (LSTM). Despite these theoretical advantages, however, unlike CNNs, previous MD-LSTM variants were hard to parallelize on GPUs. Here we re-arrange the traditional cuboid order of computations in MD-LSTM in pyramidal fashion. The resulting PyraMiD-LSTM is easy to parallelize, especially for 3D data such as stacks of brain slice images. PyraMiD-LSTM achieved best known pixel-wise brain image segmentation results on MRBrainS13 (and competitive results on EM-ISBI12).

Dual Adversarial Semantics-Consistent Network for Generalized Zero-Shot Learning

Generalized zero-shot learning (GZSL) is a challenging class of vision and knowledge transfer problems in which both seen and unseen classes appear during testing. Existing GZSL approaches either suffer from semantic loss and discard discriminative information at the embedding stage, or cannot guarantee the visual-semantic interactions. To address these limitations, we propose a Dual Adversarial Semantics-Consistent Network (referred to as DASCN), which learns both primal and dual Generative Adversarial Networks (GANs) in a unified framework for GZSL. In DASCN, the primal GAN learns to synthesize inter-class discriminative and semantics-preserving visual features from both the semantic representations of seen/unseen classes and the ones reconstructed by the dual GAN. The dual GAN enforces the synthetic visual features to represent prior semantic knowledge well via semantics-consistent adversarial learning. To the best of our knowledge, this is the first work that employs a novel dual-GAN mechanism for GZSL. Extensive experiments show that our approach achieves significant improvements over the state-of-the-art approaches.

PAC learnability versus VC dimension: a footnote to a basic result of statistical learning

A fundamental result of statistical learnig theory states that a concept class is PAC learnable if and only if it is a uniform Glivenko-Cantelli class if and only if the VC dimension of the class is finite. However, the theorem is only valid under special assumptions of measurability of the class, in which case the PAC learnability even becomes consistent. Otherwise, there is a classical example, constructed under the Continuum Hypothesis by Dudley and Durst and further adapted by Blumer, Ehrenfeucht, Haussler, and Warmuth, of a concept class of VC dimension one which is neither uniform Glivenko-Cantelli nor consistently PAC learnable. We show that, rather surprisingly, under an additional set-theoretic hypothesis which is much milder than the Continuum Hypothesis (Martin's Axiom), PAC learnability is equivalent to finite VC dimension for every concept class.

Bayesian optimization for automated model selection

Despite the success of kernel-based nonparametric methods, kernel selection still requires considerable expertise, and is often described as a “black art.” We present a sophisticated method for automatically searching for an appropriate kernel from an infinite space of potential choices. Previous efforts in this direction have focused on traversing a kernel grammar, only examining the data via computation of marginal likelihood. Our proposed search method is based on Bayesian optimization in model space, where we reason about model evidence as a function to be maximized. We explicitly reason about the data distribution and how it induces similarity between potential model choices in terms of the explanations they can offer for observed data. In this light, we construct a novel kernel between models to explain a given dataset. Our method is capable of finding a model that explains a given dataset well without any human assistance, often with fewer computations of model evidence than previous approaches, a claim we demonstrate empirically.

Nonparametric Bayesian Lomax delegate racing for survival analysis with competing risks

We propose Lomax delegate racing (LDR) to explicitly model the mechanism of survival under competing risks and to interpret how the covariates accelerate or decelerate the time to event. LDR explains non-monotonic covariate effects by racing a potentially infinite number of sub-risks, and consequently relaxes the ubiquitous proportional-hazards assumption which may be too restrictive. Moreover, LDR is naturally able to model not only censoring, but also missing event times or event types. For inference, we develop a Gibbs sampler under data augmentation for moderately sized data, along with a stochastic gradient descent maximum a posteriori inference algorithm for big data applications. Illustrative experiments are provided on both synthetic and real datasets, and comparison with various benchmark algorithms for survival analysis with competing risks demonstrates distinguished performance of LDR.

Reinforcement Learning Neural Turing Machines - Revised

The expressive power of a machine learning model is closely related to the number of sequential computational steps it can learn. For example, Deep Neural Networks have been more successful than shallow networks because they can perform a greater number of sequential computational steps (each highly parallel). The Neural Turing Machine (NTM) is a model that can compactly express an even greater number of sequential computational steps, so it is even more powerful than a DNN. Its memory addressing operations are designed to be differentiable; thus the NTM can be trained with backpropagation.

How Many Folders Do You Really Need?

Email classification is still a mostly manual task. Consequently, most Web mail users never define a single folder. Recently however, automatic classification offering the same categories to all users has started to appear in some Web mail clients, such as AOL or Gmail. We adopt this approach, rather than previous (unsuccessful) personalized approaches because of the change in the nature of consumer email traffic, which is now dominated by (non-spam) machine-generated email. We propose here a novel approach for (1) automatically distinguishing between personal and machine-generated email and (2) classifying messages into latent categories, without requiring users to have defined any folder. We report how we have discovered that a set of 6 "latent" categories (one for human- and the others for machine-generated messages) can explain a significant portion of email traffic. We describe in details the steps involved in building a Web-scale email categorization system, from the collection of ground-truth labels, the selection of features to the training of models. Experimental evaluation was performed on more than 500 billion messages received during a period of six months by users of Yahoo mail service, who elected to be part of such research studies. Our system achieved precision and recall rates close to 90% and the latent categories we discovered were shown to cover 70% of both email traffic and email search queries. We believe that these results pave the way for a change of approach in the Web mail industry, and could support the invention of new large-scale email discovery paradigms that had not been possible before.

TwiSE at SemEval-2016 Task 4: Twitter Sentiment Classification

This paper describes the participation of the team "TwiSE" in the SemEval 2016 challenge. Specifically, we participated in Task 4, namely "Sentiment Analysis in Twitter" for which we implemented sentiment classification systems for subtasks A, B, C and D. Our approach consists of two steps. In the first step, we generate and validate diverse feature sets for twitter sentiment evaluation, inspired by the work of participants of previous editions of such challenges. In the second step, we focus on the optimization of the evaluation measures of the different subtasks. To this end, we examine different learning strategies by validating them on the data provided by the task organisers. For our final submissions we used an ensemble learning approach (stacked generalization) for Subtask A and single linear models for the rest of the subtasks. In the official leaderboard we were ranked 9/35, 8/19, 1/11 and 2/14 for subtasks A, B, C and D respectively.\footnote{We make the code available for research purposes at \url{

Multiscale probability transformation of basic probability assignment

Decision making is still an open issue in the application of Dempster-Shafer evidence theory. A lot of works have been presented for it. In the transferable belief model (TBM), pignistic probabilities based on the basic probability as- signments are used for decision making. In this paper, multiscale probability transformation of basic probability assignment based on the belief function and the plausibility function is proposed, which is a generalization of the pignistic probability transformation. In the multiscale probability function, a factor q based on the Tsallis entropy is used to make the multiscale prob- abilities diversified. An example is shown that the multiscale probability transformation is more reasonable in the decision making.

Ambiguity-Driven Fuzzy C-Means Clustering: How to Detect Uncertain Clustered Records

As a well-known clustering algorithm, Fuzzy C-Means (FCM) allows each input sample to belong to more than one cluster, providing more flexibility than non-fuzzy clustering methods. However, the accuracy of FCM is subject to false detections caused by noisy records, weak feature selection and low certainty of the algorithm in some cases. The false detections are very important in some decision-making application domains like network security and medical diagnosis, where weak decisions based on such false detections may lead to catastrophic outcomes. They are mainly emerged from making decisions about a subset of records that do not provide enough evidence to make a good decision. In this paper, we propose a method for detecting such ambiguous records in FCM by introducing a certainty factor to decrease invalid detections. This approach enables us to send the detected ambiguous records to another discrimination method for a deeper investigation, thus increasing the accuracy by lowering the error rate. Most of the records are still processed quickly and with low error rate which prevents performance loss compared to similar hybrid methods. Experimental results of applying the proposed method on several datasets from different domains show a significant decrease in error rate as well as improved sensitivity of the algorithm.

An MLP based Approach for Recognition of Handwritten `Bangla' Numerals

The work presented here involves the design of a Multi Layer Perceptron (MLP) based pattern classifier for recognition of handwritten Bangla digits using a 76 element feature vector. Bangla is the second most popular script and language in the Indian subcontinent and the fifth most popular language in the world. The feature set developed for representing handwritten Bangla numerals here includes 24 shadow features, 16 centroid features and 36 longest-run features. On experimentation with a database of 6000 samples, the technique yields an average recognition rate of 96.67% evaluated after three-fold cross validation of results. It is useful for applications related to OCR of handwritten Bangla Digit and can also be extended to include OCR of handwritten characters of Bangla alphabet.

The Computational Theory of Intelligence: Data Aggregation

In this paper, we will expound upon the concepts proffered in [1], where we proposed an information theoretic approach to intelligence in the computational sense. We will examine data and meme aggregation, and study the effect of limited resources on the resulting meme amplitudes.

Optimal Rates for Random Fourier Features

Kernel methods represent one of the most powerful tools in machine learning to tackle problems expressed in terms of function values and derivatives due to their capability to represent and model complex relations. While these methods show good versatility, they are computationally intensive and have poor scalability to large data as they require operations on Gram matrices. In order to mitigate this serious computational limitation, recently randomized constructions have been proposed in the literature, which allow the application of fast linear algorithms. Random Fourier features (RFF) are among the most popular and widely applied constructions: they provide an easily computable, low-dimensional feature representation for shift-invariant kernels. Despite the popularity of RFFs, very little is understood theoretically about their approximation quality. In this paper, we provide the first detailed theoretical analysis about the approximation quality of RFFs by establishing optimal (in terms of the RFF dimension) performance guarantees in uniform and $L^r$ ($1\le r&lt;\infty$) norms. We also propose a RFF approximation to derivatives of a kernel with a theoretical study on its approximation quality.

Generalization of ERM in Stochastic Convex Optimization: The Dimension Strikes Back

In stochastic convex optimization the goal is to minimize a convex function $F(x) \doteq {\mathbf E}_{{\mathbf f}\sim D}[{\mathbf f}(x)]$ over a convex set $\cal K \subset {\mathbb R}^d$ where $D$ is some unknown distribution and each $f(\cdot)$ in the support of $D$ is convex over $\cal K$. The optimization is commonly based on i.i.d.~samples $f^1,f^2,\ldots,f^n$ from $D$. A standard approach to such problems is empirical risk minimization (ERM) that optimizes $F_S(x) \doteq \frac{1}{n}\sum_{i\leq n} f^i(x)$. Here we consider the question of how many samples are necessary for ERM to succeed and the closely related question of uniform convergence of $F_S$ to $F$ over $\cal K$. We demonstrate that in the standard $\ell_p/\ell_q$ setting of Lipschitz-bounded functions over a $\cal K$ of bounded radius, ERM requires sample size that scales linearly with the dimension $d$. This nearly matches standard upper bounds and improves on $\Omega(\log d)$ dependence proved for $\ell_2/\ell_2$ setting by Shalev-Shwartz et al. (2009). In stark contrast, these problems can be solved using dimension-independent number of samples for $\ell_2/\ell_2$ setting and $\log d$ dependence for $\ell_1/\ell_\infty$ setting using other approaches. We also demonstrate that for a more general class of range-bounded (but not Lipschitz-bounded) stochastic convex programs an even stronger gap appears already in dimension 2.

VEEGAN: Reducing Mode Collapse in GANs using Implicit Variational Learning

Deep generative models provide powerful tools for distributions over complicated manifolds, such as those of natural images. But many of these methods, including generative adversarial networks (GANs), can be difficult to train, in part because they are prone to mode collapse, which means that they characterize only a few modes of the true distribution. To address this, we introduce VEEGAN, which features a reconstructor network, reversing the action of the generator by mapping from data to noise. Our training objective retains the original asymptotic consistency guarantee of GANs, and can be interpreted as a novel autoencoder loss over the noise. In sharp contrast to a traditional autoencoder over data points, VEEGAN does not require specifying a loss function over the data, but rather only over the representations, which are standard normal by assumption. On an extensive set of synthetic and real world image datasets, VEEGAN indeed resists mode collapsing to a far greater extent than other recent GAN variants, and produces more realistic samples.

Information-Propogation-Enhanced Neural Machine Translation by Relation Model

Even though sequence-to-sequence neural machine translation (NMT) model have achieved state-of-art performance in the recent fewer years, but it is widely concerned that the recurrent neural network (RNN) units are very hard to capture the long-distance state information, which means RNN can hardly find the feature with long term dependency as the sequence becomes longer. Similarly, convolutional neural network (CNN) is introduced into NMT for speeding recently, however, CNN focus on capturing the local feature of the sequence; To relieve this issue, we incorporate a relation network into the standard encoder-decoder framework to enhance information-propogation in neural network, ensuring that the information of the source sentence can flow into the decoder adequately. Experiments show that proposed framework outperforms the statistical MT model and the state-of-art NMT model significantly on two data sets with different scales.

Scenario Submodular Cover

Many problems in Machine Learning can be modeled as submodular optimization problems. Recent work has focused on stochastic or adaptive versions of these problems. We consider the Scenario Submodular Cover problem, which is a counterpart to the Stochastic Submodular Cover problem studied by Golovin and Krause. In Scenario Submodular Cover, the goal is to produce a cover with minimum expected cost, where the expectation is with respect to an empirical joint distribution, given as input by a weighted sample of realizations. In contrast, in Stochastic Submodular Cover, the variables of the input distribution are assumed to be independent, and the distribution of each variable is given as input. Building on algorithms developed by Cicalese et al. and Golovin and Krause for related problems, we give two approximation algorithms for Scenario Submodular Cover over discrete distributions. The first achieves an approximation factor of O(log Qm), where m is the size of the sample and Q is the goal utility. The second, simpler algorithm achieves an approximation bound of O(log QW), where Q is the goal utility and W is the sum of the integer weights. (Both bounds assume an integer-valued utility function.) Our results yield approximation bounds for other problems involving non-independent distributions that are explicitly specified by their support.

Machine Learned Learning Machines

There are two common approaches for optimizing the performance of a machine: genetic algorithms and machine learning. A genetic algorithm is applied over many generations whereas machine learning works by applying feedback until the system meets a performance threshold. Though these are methods that typically operate separately, we combine evolutionary adaptation and machine learning into one approach. Our focus is on machines that can learn during their lifetime, but instead of equipping them with a machine learning algorithm we aim to let them evolve their ability to learn by themselves. We use evolvable networks of probabilistic and deterministic logic gates, known as Markov Brains, as our computational model organism. The ability of Markov Brains to learn is augmented by a novel adaptive component that can change its computational behavior based on feedback. We show that Markov Brains can indeed evolve to incorporate these feedback gates to improve their adaptability to variable environments. By combining these two methods, we now also implemented a computational model that can be used to study the evolution of learning.

On the Curved Geometry of Accelerated Optimization

In this work we propose a differential geometric motivation for Nesterov's accelerated gradient method (AGM) for strongly-convex problems. By considering the optimization procedure as occurring on a Riemannian manifold with a natural structure, The AGM method can be seen as the proximal point method applied in this curved space. This viewpoint can also be extended to the continuous time case, where the accelerated gradient method arises from the natural block-implicit Euler discretization of an ODE on the manifold. We provide an analysis of the convergence rate of this ODE for quadratic objectives.

A Mixed Graphical Model for Rhythmic Parsing

A method is presented for the rhythmic parsing problem: Given a sequence of observed musical note onset times, we estimate the corresponding notated rhythm and tempo process. A graphical model is developed that represents the simultaneous evolution of tempo and rhythm and relates these hidden quantities to observations. The rhythm variables are discrete and the tempo and observation variables are continuous. We show how to compute the globally most likely configuration of the tempo and rhythm variables given an observation of note onset times. Preliminary experiments are presented on a small data set. A generalization to arbitrary conditional Gaussian distributions is outlined.

The Perceptron with Dynamic Margin

The classical perceptron rule provides a varying upper bound on the maximum margin, namely the length of the current weight vector divided by the total number of updates up to that time. Requiring that the perceptron updates its internal state whenever the normalized margin of a pattern is found not to exceed a certain fraction of this dynamic upper bound we construct a new approximate maximum margin classifier called the perceptron with dynamic margin (PDM). We demonstrate that PDM converges in a finite number of steps and derive an upper bound on them. We also compare experimentally PDM with other perceptron-like algorithms and support vector machines on hard margin tasks involving linear kernels which are equivalent to 2-norm soft margin.

DROW: Real-Time Deep Learning based Wheelchair Detection in 2D Range Data

We introduce the DROW detector, a deep learning based detector for 2D range data. Laser scanners are lighting invariant, provide accurate range data, and typically cover a large field of view, making them interesting sensors for robotics applications. So far, research on detection in laser range data has been dominated by handcrafted features and boosted classifiers, potentially losing performance due to suboptimal design choices. We propose a Convolutional Neural Network (CNN) based detector for this task. We show how to effectively apply CNNs for detection in 2D range data, and propose a depth preprocessing step and voting scheme that significantly improve CNN performance. We demonstrate our approach on wheelchairs and walkers, obtaining state of the art detection results. Apart from the training data, none of our design choices limits the detector to these two classes, though. We provide a ROS node for our detector and release our dataset containing 464k laser scans, out of which 24k were annotated for training.

Shallow Discourse Parsing Using Distributed Argument Representations and Bayesian Optimization

This paper describes the Georgia Tech team's approach to the CoNLL-2016 supplementary evaluation on discourse relation sense classification. We use long short-term memories (LSTM) to induce distributed representations of each argument, and then combine these representations with surface features in a neural network. The architecture of the neural network is determined by Bayesian hyperparameter search.

Hybrid Dialog State Tracker with ASR Features

This paper presents a hybrid dialog state tracker enhanced by trainable Spoken Language Understanding (SLU) for slot-filling dialog systems. Our architecture is inspired by previously proposed neural-network-based belief-tracking systems. In addition, we extended some parts of our modular architecture with differentiable rules to allow end-to-end training. We hypothesize that these rules allow our tracker to generalize better than pure machine-learning based systems. For evaluation, we used the Dialog State Tracking Challenge (DSTC) 2 dataset - a popular belief tracking testbed with dialogs from restaurant information system. To our knowledge, our hybrid tracker sets a new state-of-the-art result in three out of four categories within the DSTC2.

Exploiting Numerical Sparsity for Efficient Learning : Faster Eigenvector Computation and Regression

In this paper, we obtain improved running times for regression and top eigenvector computation for numerically sparse matrices. Given a data matrix $\mat{A} \in \R^{n \times d}$ where every row $a \in \R^d$ has $\|a\|_2^2 \leq L$ and numerical sparsity $\leq s$, i.e. $\|a\|_1^2 / \|a\|_2^2 \leq s$, we provide faster algorithms for these problems for many parameter settings. For top eigenvector computation, when $\gap > 0$ is the relative gap between the top two eigenvectors of $\mat{A}^\top \mat{A}$ and $r$ is the stable rank of $\mat{A}$ we obtain a running time of $\otilde(nd + r(s + \sqrt{r s}) / \gap^2)$ improving upon the previous best unaccelerated running time of $O(nd + r d / \gap^2)$. As $r \leq d$ and $s \leq d$ our algorithm everywhere improves or matches the previous bounds for all parameter settings. For regression, when $\mu > 0$ is the smallest eigenvalue of $\mat{A}^\top \mat{A}$ we obtain a running time of $\otilde(nd + (nL / \mu) \sqrt{s nL / \mu})$ improving upon the previous best unaccelerated running time of $\otilde(nd + n L d / \mu)$. This result expands when regression can be solved in nearly linear time from when $L/\mu = \otilde(1)$ to when $L / \mu = \otilde(d^{2/3} / (sn)^{1/3})$. Furthermore, we obtain similar improvements even when row norms and numerical sparsities are non-uniform and we show how to achieve even faster running times by accelerating using approximate proximal point \cite{frostig2015regularizing} / catalyst \cite{lin2015universal}. Our running times depend only on the size of the input and natural numerical measures of the matrix, i.e. eigenvalues and $\ell_p$ norms, making progress on a key open problem regarding optimal running times for efficient large-scale learning.

MidiNet: A Convolutional Generative Adversarial Network for Symbolic-domain Music Generation

In this paper, we present MidiNet, a deep convolutional neural network (CNN) based generative adversarial network (GAN) that is intended to provide a general, highly adaptive network structure for symbolic-domain music generation. The network takes random noise as input and generates a melody sequence one mea- sure (bar) after another. Moreover, it has a novel reflective CNN sub-model that allows us to guide the generation process by providing not only 1D but also 2D conditions. In our implementation, we used the intended chord of the current bar as a 1D condition to provide a harmonic context, and the melody generated for the preceding bar previously as a 2D condition to provide sequential information. The output of the network is a 16 by 128 matrix each time, representing the presence of each of the 128 MIDI notes in the generated melody sequence of that bar, with the smallest temporal unit being the sixteenth note. MidiNet can generate music of arbitrary number of bars, by concatenating these 16 by 128 matrices. The melody sequence can then be played back with a synthesizer. We provide example clips showing the effectiveness of MidiNet in generating harmonic music.

Online Adaptive Methods, Universality and Acceleration

We present a novel method for convex unconstrained optimization that, without any modifications ensures: (1) accelerated convergence rate for smooth objectives, (2) standard convergence rate in the general (non-smooth) setting, and (3) standard convergence rate in the stochastic optimization setting.<br /> To the best of our knowledge, this is the first method that simultaneously applies to all of the above settings.<br /> At the heart of our method is an adaptive learning rate rule that employs importance weights, in the spirit of adaptive online learning algorithms [duchi2011adaptive,levy2017online], combined with an update that linearly couples two sequences, in the spirit of [AllenOrecchia2017]. An empirical examination of our method demonstrates its applicability to the above mentioned scenarios and corroborates our theoretical findings.

Constructing Fast Network through Deconstruction of Convolution

Convolutional neural networks have achieved great success in various vision tasks; however, they incur heavy resource costs. By using deeper and wider networks, network accuracy can be improved rapidly. However, in an environment with limited resources (e.g., mobile applications), heavy networks may not be usable. This study shows that naive convolution can be deconstructed into a shift operation and pointwise convolution. To cope with various convolutions, we propose a new shift operation called active shift layer (ASL) that formulates the amount of shift as a learnable function with shift parameters. This new layer can be optimized end-to-end through backpropagation and it can provide optimal shift values. Finally, we apply this layer to a light and fast network that surpasses existing state-of-the-art networks.

Near-Optimal Active Learning of Halfspaces via Query Synthesis in the Noisy Setting

In this paper, we consider the problem of actively learning a linear classifier through query synthesis where the learner can construct artificial queries in order to estimate the true decision boundaries. This problem has recently gained a lot of interest in automated science and adversarial reverse engineering for which only heuristic algorithms are known. In such applications, queries can be constructed de novo to elicit information (e.g., automated science) or to evade detection with minimal cost (e.g., adversarial reverse engineering).

GOOWE: Geometrically Optimum and Online-Weighted Ensemble Classifier for Evolving Data Streams

Designing adaptive classifiers for an evolving data stream is a challenging task due to the data size and its dynamically changing nature. Combining individual classifiers in an online setting, the ensemble approach, is a well-known solution. It is possible that a subset of classifiers in the ensemble outperforms others in a time-varying fashion. However, optimum weight assignment for component classifiers is a problem which is not yet fully addressed in online evolving environments. We propose a novel data stream ensemble classifier, called Geometrically Optimum and Online-Weighted Ensemble (GOOWE), which assigns optimum weights to the component classifiers using a sliding window containing the most recent data instances. We map vote scores of individual classifiers and true class labels into a spatial environment. Based on the Euclidean distance between vote scores and ideal-points, and using the linear least squares (LSQ) solution, we present a novel, dynamic, and online weighting approach. While LSQ is used for batch mode ensemble classifiers, it is the first time that we adapt and use it for online environments by providing a spatial modeling of online ensembles. In order to show the robustness of the proposed algorithm, we use real-world datasets and synthetic data generators using the MOA libraries. First, we analyze the impact of our weighting system on prediction accuracy through two scenarios. Second, we compare GOOWE with 8 state-of-the-art ensemble classifiers in a comprehensive experimental environment. Our experiments show that GOOWE provides improved reactions to different types of concept drift compared to our baselines. The statistical tests indicate a significant improvement in accuracy, with conservative time and memory requirements.

Geometric Descent Method for Convex Composite Minimization

In this paper, we extend the geometric descent method recently proposed by Bubeck, Lee and Singh to tackle nonsmooth and strongly convex composite problems. We prove that our proposed algorithm, dubbed geometric proximal gradient method (GeoPG), converges with a linear rate $(1-1/\sqrt{\kappa})$ and thus achieves the optimal rate among first-order methods, where $\kappa$ is the condition number of the problem. Numerical results on linear regression and logistic regression with elastic net regularization show that GeoPG compares favorably with Nesterov's accelerated proximal gradient method, especially when the problem is ill-conditioned.

Universal and Determined Constructors of Multisets of Objects

This paper contains analysis of creation of sets and multisets as an approach for modeling of some aspects of human thinking. The creation of sets is considered within constructive object-oriented version of set theory (COOST), from different sides, in particular classical set theory, object-oriented programming (OOP) and development of intelligent information systems (IIS). The main feature of COOST in contrast to other versions of set theory is an opportunity to describe essences of objects more precisely, using their properties and methods, which can be applied to them. That is why this version of set theory is object-oriented and close to OOP. Within COOST, the author proposes universal constructor of multisets of objects that gives us a possibility to create arbitrary multisets of objects. In addition, a few determined constructors of multisets of objects, which allow creating multisets, using strictly defined schemas, also are proposed in the paper. Such constructors are very useful in cases of very big cardinalities of multisets, because they give us an opportunity to calculate a multiplicity of each object and cardinality of multiset before its creation. The proposed constructors of multisets of objects allow us to model in a sense corresponding processes of human thought, that in turn give us an opportunity to develop IIS, using these tools.

Robust Order Scheduling in the Fashion Industry: A Multi-Objective Optimization Approach

In the fashion industry, order scheduling focuses on the assignment of production orders to appropriate production lines. In reality, before a new order can be put into production, a series of activities known as pre-production events need to be completed. In addition, in real production process, owing to various uncertainties, the daily production quantity of each order is not always as expected. In this research, by considering the pre-production events and the uncertainties in the daily production quantity, robust order scheduling problems in the fashion industry are investigated with the aid of a multi-objective evolutionary algorithm (MOEA) called nondominated sorting adaptive differential evolution (NSJADE). The experimental results illustrate that it is of paramount importance to consider pre-production events in order scheduling problems in the fashion industry. We also unveil that the existence of the uncertainties in the daily production quantity heavily affects the order scheduling.

Optimizing Gross Merchandise Volume via DNN-MAB Dynamic Ranking Paradigm

With the transition from people's traditional `brick-and-mortar' shopping to online mobile shopping patterns in web 2.0 $\mathit{era}$, the recommender system plays a critical role in E-Commerce and E-Retails. This is especially true when designing this system for more than $\mathbf{236~million}$ daily active users. Ranking strategy, the key module of the recommender system, needs to be precise, accurate, and responsive for estimating customers' intents. We propose a dynamic ranking paradigm, named as DNN-MAB, that is composed of a pairwise deep neural network (DNN) $\mathit{pre}$-ranker connecting a revised multi-armed bandit (MAB) dynamic $\mathit{post}$-ranker. By taking into account of explicit and implicit user feedbacks such as impressions, clicks, conversions, etc. DNN-MAB is able to adjust DNN $\mathit{pre}$-ranking scores to assist customers locating items they are interested in most so that they can converge quickly and frequently. To the best of our knowledge, frameworks like DNN-MAB have not been discussed in the previous literature to either E-Commerce or machine learning audiences. In practice, DNN-MAB has been deployed to production and it easily outperforms against other state-of-the-art models by significantly lifting the gross merchandise volume (GMV) which is the objective metrics at JD.

Reinforcement Learning with Convex Constraints

In standard reinforcement learning (RL), a learning agent seeks to optimize the overall reward. However, many key aspects of a desired behavior are more naturally expressed as constraints. For instance, the designer may want to limit the use of unsafe actions, increase the diversity of trajectories to enable exploration, or approximate expert trajectories when rewards are sparse. In this paper, we propose an algorithmic scheme that can handle a wide class of constraints in RL tasks: specifically, any constraints that require expected values of some vector measurements (such as the use of an action) to lie in a convex set. This captures previously studied constraints (such as safety and proximity to an expert), but also enables new classes of constraints (such as diversity). Our approach comes with rigorous theoretical guarantees and only relies on the ability to approximately solve standard RL tasks. As a result, it can be easily adapted to work with any model-free or model-based RL. In our experiments, we show that it matches previous algorithms that enforce safety via constraints, but can also enforce new properties that these algorithms do not incorporate, such as diversity.

Exponentially Weighted Imitation Learning for Batched Historical Data

We consider deep policy learning with only batched historical trajectories. The main challenge of this problem is that the learner no longer has a simulator or ``environment oracle'' as in most reinforcement learning settings. To solve this problem, we propose a monotonic advantage reweighted imitation learning strategy that is applicable to problems with complex nonlinear function approximation and works well with hybrid (discrete and continuous) action space. The method does not rely on the knowledge of the behavior policy, thus can be used to learn from data generated by an unknown policy. Under mild conditions, our algorithm, though surprisingly simple, has a policy improvement bound and outperforms most competing methods empirically. Thorough numerical results are also provided to demonstrate the efficacy of the proposed methodology.

Revisiting Batch Normalization For Practical Domain Adaptation

Deep neural networks (DNN) have shown unprecedented success in various computer vision applications such as image classification and object detection. However, it is still a common (yet inconvenient) practice to prepare at least tens of thousands of labeled image to fine-tune a network on every task before the model is ready to use. Recent study shows that a DNN has strong dependency towards the training dataset, and the learned features cannot be easily transferred to a different but relevant task without fine-tuning.

Understanding and Improving Layer Normalization

Layer normalization (LayerNorm) is a technique to normalize the distributions of intermediate layers. It enables smoother gradients, faster training, and better generalization accuracy. However, it is still unclear where the effectiveness stems from. In this paper, our main contribution is to take a step further in understanding LayerNorm. Many of previous studies believe that the success of LayerNorm comes from forward normalization. Unlike them, we find that the derivatives of the mean and variance are more important than forward normalization by re-centering and re-scaling backward gradients. Furthermore, we find that the parameters of LayerNorm, including the bias and gain, increase the risk of over-fitting and do not work in most cases. Experiments show that a simple version of LayerNorm (LayerNorm-simple) without the bias and gain outperforms LayerNorm on four datasets. It obtains the state-of-the-art performance on En-Vi machine translation. To address the over-fitting problem, we propose a new normalization method, Adaptive Normalization (AdaNorm), by replacing the bias and gain with a new transformation function. Experiments show that AdaNorm demonstrates better results than LayerNorm on seven out of eight datasets.

"Pale as death" or "p\^ale comme la mort" : Frozen similes used as literary clich\'es

The present study is focused on the automatic identification and description of frozen similes in British and French novels written between the 19 th century and the beginning of the 20 th century. Two main patterns of frozen similes were considered: adjectival ground + simile marker + nominal vehicle (e.g. happy as a lark) and eventuality + simile marker + nominal vehicle (e.g. sleep like a top). All potential similes and their components were first extracted using a rule-based algorithm. Then, frozen similes were identified based on reference lists of existing similes and semantic distance between the tenor and the vehicle. The results obtained tend to confirm the fact that frozen similes are not used haphazardly in literary texts. In addition, contrary to how they are often presented, frozen similes often go beyond the ground or the eventuality and the vehicle to also include the tenor.

Large-Scale Price Optimization via Network Flow

This paper deals with price optimization, which is to find the best pricing strategy that maximizes revenue or profit, on the basis of demand forecasting models. Though recent advances in regression technologies have made it possible to reveal price-demand relationship of a number of multiple products, most existing price optimization methods, such as mixed integer programming formulation, cannot handle tens or hundreds of products because of their high computational costs. To cope with this problem, this paper proposes a novel approach based on network flow algorithms. We reveal a connection between supermodularity of the revenue and cross elasticity of demand. On the basis of this connection, we propose an efficient algorithm that employs network flow algorithms. The proposed algorithm can handle hundreds or thousands of products, and returns an exact optimal solution under an assumption regarding cross elasticity of demand. Even in case in which the assumption does not hold, the proposed algorithm can efficiently find approximate solutions as good as can other state-of-the-art methods, as empirical results show.

Similarity Measures on Preference Structures, Part II: Utility Functions

In previous work cite{Ha98:Towards} we presented a case-based approach to eliciting and reasoning with preferences. A key issue in this approach is the definition of similarity between user preferences. We introduced the probabilistic distance as a measure of similarity on user preferences, and provided an algorithm to compute the distance between two partially specified {em value} functions. This is for the case of decision making under {em certainty}. In this paper we address the more challenging issue of computing the probabilistic distance in the case of decision making under{em uncertainty}. We provide an algorithm to compute the probabilistic distance between two partially specified {em utility} functions. We demonstrate the use of this algorithm with a medical data set of partially specified patient preferences,where none of the other existing distancemeasures appear definable. Using this data set, we also demonstrate that the case-based approach to preference elicitation isapplicable in domains with uncertainty. Finally, we provide a comprehensive analytical comparison of the probabilistic distance with some existing distance measures on preferences.

DASA: Domain Adaptation in Stacked Autoencoders using Systematic Dropout

Domain adaptation deals with adapting behaviour of machine learning based systems trained using samples in source domain to their deployment in target domain where the statistics of samples in both domains are dissimilar. The task of directly training or adapting a learner in the target domain is challenged by lack of abundant labeled samples. In this paper we propose a technique for domain adaptation in stacked autoencoder (SAE) based deep neural networks (DNN) performed in two stages: (i) unsupervised weight adaptation using systematic dropouts in mini-batch training, (ii) supervised fine-tuning with limited number of labeled samples in target domain. We experimentally evaluate performance in the problem of retinal vessel segmentation where the SAE-DNN is trained using large number of labeled samples in the source domain (DRIVE dataset) and adapted using less number of labeled samples in target domain (STARE dataset). The performance of SAE-DNN measured using $logloss$ in source domain is $0.19$, without and with adaptation are $0.40$ and $0.18$, and $0.39$ when trained exclusively with limited samples in target domain. The area under ROC curve is observed respectively as $0.90$, $0.86$, $0.92$ and $0.87$. The high efficiency of vessel segmentation with DASA strongly substantiates our claim.

Recurrently Controlled Recurrent Networks

Recurrent neural networks (RNNs) such as long short-term memory and gated recurrent units are pivotal building blocks across a broad spectrum of sequence modeling problems. This paper proposes a recurrently controlled recurrent network (RCRN) for expressive and powerful sequence encoding. More concretely, the key idea behind our approach is to learn the recurrent gating functions using recurrent networks. Our architecture is split into two components - a controller cell and a listener cell whereby the recurrent controller actively influences the compositionality of the listener cell. We conduct extensive experiments on a myriad of tasks in the NLP domain such as sentiment analysis (SST, IMDb, Amazon reviews, etc.), question classification (TREC), entailment classification (SNLI, SciTail), answer selection (WikiQA, TrecQA) and reading comprehension (NarrativeQA). Across all 26 datasets, our results demonstrate that RCRN not only consistently outperforms BiLSTMs but also stacked BiLSTMs, suggesting that our controller architecture might be a suitable replacement for the widely adopted stacked architecture. Additionally, RCRN achieves state-of-the-art results on several well-established datasets.

On Stochastic Belief Revision and Update and their Combination

I propose a framework for an agent to change its probabilistic beliefs when a new piece of propositional information $\alpha$ is observed. Traditionally, belief change occurs by either a revision process or by an update process, depending on whether the agent is informed with $\alpha$ in a static world or, respectively, whether $\alpha$ is a 'signal' from the environment due to an event occurring. Boutilier suggested a unified model of qualitative belief change, which "combines aspects of revision and update, providing a more realistic characterization of belief change." In this paper, I propose a unified model of quantitative belief change, where an agent's beliefs are represented as a probability distribution over possible worlds. As does Boutilier, I take a dynamical systems perspective. The proposed approach is evaluated against several rationality postulated, and some properties of the approach are worked out.

Efficient Representations for Life-Long Learning and Autoencoding

It has been a long-standing goal in machine learning, as well as in AI more generally, to develop life-long learning systems that learn many different tasks over time, and reuse insights from tasks learned, "learning to learn" as they do so. In this work we pose and provide efficient algorithms for several natural theoretical formulations of this goal. Specifically, we consider the problem of learning many different target functions over time, that share certain commonalities that are initially unknown to the learning algorithm. Our aim is to learn new internal representations as the algorithm learns new target functions, that capture this commonality and allow subsequent learning tasks to be solved more efficiently and from less data. We develop efficient algorithms for two very different kinds of commonalities that target functions might share: one based on learning common low-dimensional and unions of low-dimensional subspaces and one based on learning nonlinear Boolean combinations of features. Our algorithms for learning Boolean feature combinations additionally have a dual interpretation, and can be viewed as giving an efficient procedure for constructing near-optimal sparse Boolean autoencoders under a natural "anchor-set" assumption.

Learning to Discover Efficient Mathematical Identities

In this paper we explore how machine learning techniques can be applied to the discovery of efficient mathematical identities. We introduce an attribute grammar framework for representing symbolic expressions. Given a set of grammar rules we build trees that combine different rules, looking for branches which yield compositions that are analytically equivalent to a target expression, but of lower computational complexity. However, as the size of the trees grows exponentially with the complexity of the target expression, brute force search is impractical for all but the simplest of expressions. Consequently, we introduce two novel learning approaches that are able to learn from simpler expressions to guide the tree search. The first of these is a simple n-gram model, the other being a recursive neural-network. We show how these approaches enable us to derive complex identities, beyond reach of brute-force search, or human derivation.

Target-Side Context for Discriminative Models in Statistical Machine Translation

Discriminative translation models utilizing source context have been shown to help statistical machine translation performance. We propose a novel extension of this work using target context information. Surprisingly, we show that this model can be efficiently integrated directly in the decoding process. Our approach scales to large training data sizes and results in consistent improvements in translation quality on four language pairs. We also provide an analysis comparing the strengths of the baseline source-context model with our extended source-context and target-context model and we show that our extension allows us to better capture morphological coherence. Our work is freely available as part of Moses.

Next Generation Business Intelligence and Analytics: A Survey

Business Intelligence and Analytics (BI&amp;A) is the process of extracting and predicting business-critical insights from data. Traditional BI focused on data collection, extraction, and organization to enable efficient query processing for deriving insights from historical data. With the rise of big data and cloud computing, there are many challenges and opportunities for the BI. Especially with the growing number of data sources, traditional BI\&amp;A are evolving to provide intelligence at different scales and perspectives - operational BI, situational BI, self-service BI. In this survey, we review the evolution of business intelligence systems in full scale from back-end architecture to and front-end applications. We focus on the changes in the back-end architecture that deals with the collection and organization of the data. We also review the changes in the front-end applications, where analytic services and visualization are the core components. Using a uses case from BI in Healthcare, which is one of the most complex enterprises, we show how BI\&amp;A will play an important role beyond the traditional usage. The survey provides a holistic view of Business Intelligence and Analytics for anyone interested in getting a complete picture of the different pieces in the emerging next generation BI\&amp;A solutions.

Training a Subsampling Mechanism in Expectation

We describe a mechanism for subsampling sequences and show how to compute its expected output so that it can be trained with standard backpropagation. We test this approach on a simple toy problem and discuss its shortcomings.

Feature Selection for Value Function Approximation Using Bayesian Model Selection

Feature selection in reinforcement learning (RL), i.e. choosing basis functions such that useful approximations of the unkown value function can be obtained, is one of the main challenges in scaling RL to real-world applications. Here we consider the Gaussian process based framework GPTD for approximate policy evaluation, and propose feature selection through marginal likelihood optimization of the associated hyperparameters. Our approach has two appealing benefits: (1) given just sample transitions, we can solve the policy evaluation problem fully automatically (without looking at the learning task, and, in theory, independent of the dimensionality of the state space), and (2) model selection allows us to consider more sophisticated kernels, which in turn enable us to identify relevant subspaces and eliminate irrelevant state variables such that we can achieve substantial computational savings and improved prediction performance.

Determination of the Internet Anonymity Influence on the Level of Aggression and Usage of Obscene Lexis

This article deals with the analysis of the semantic content of the anonymous Russian-speaking forum 2ch.hk, different verbal means of expressing of the emotional state of aggression are revealed for this site, and aggression is classified by its directions. The lexis of different Russian-and English- speaking anonymous forums (2ch.hk and iichan.hk, 4chan.org) and public community "MDK" of the Russian-speaking social network VK is analyzed and compared with the Open Corpus of the Russian language (Opencorpora.org and Brown corpus). The analysis shows that anonymity has no influence on the amount of invective items usage. The effectiveness of moderation was shown for anonymous forums. It was established that Russian obscene lexis was used to express the emotional state of aggression only in 60.4% of cases for 2ch.hk. These preliminary results show that the Russian obscene lexis on the Internet does not have direct dependence on the emotional state of aggression.

Decomposable Submodular Function Minimization: Discrete and Continuous

This paper investigates connections between discrete and continuous approaches for decomposable submodular function minimization. We provide improved running time estimates for the state-of-the-art continuous algorithms for the problem using combinatorial arguments. We also provide a systematic experimental comparison of the two types of methods, based on a clear distinction between level-0 and level-1 algorithms.

Inferring Informational Goals from Free-Text Queries: A Bayesian Approach

People using consumer software applications typically do not use technical jargon when querying an online database of help topics. Rather, they attempt to communicate their goals with common words and phrases that describe software functionality in terms of structure and objects they understand. We describe a Bayesian approach to modeling the relationship between words in a user's query for assistance and the informational goals of the user. After reviewing the general method, we describe several extensions that center on integrating additional distinctions and structure about language usage and user goals into the Bayesian models.

Annotated English

This document presents Annotated English, a system of diacritical symbols which turns English pronunciation into a precise and unambiguous process. The annotations are defined and located in such a way that the original English text is not altered (not even a letter), thus allowing for a consistent reading and learning of the English language with and without annotations. The annotations are based on a set of general rules that make the frequency of annotations not dramatically high. This makes the reader easily associate annotations with exceptions, and makes it possible to shape, internalise and consolidate some rules for the English language which otherwise are weakened by the enormous amount of exceptions in English pronunciation. The advantages of this annotation system are manifold. Any existing text can be annotated without a significant increase in size. This means that we can get an annotated version of any document or book with the same number of pages and fontsize. Since no letter is affected, the text can be perfectly read by a person who does not know the annotation rules, since annotations can be simply ignored. The annotations are based on a set of rules which can be progressively learned and recognised, even in cases where the reader has no access or time to read the rules. This means that a reader can understand most of the annotations after reading a few pages of Annotated English, and can take advantage from that knowledge for any other annotated document she may read in the future.

Galaxy-X: A Novel Approach for Multi-class Classification in an Open Universe

Classification is a fundamental task in machine learning and artificial intelligence. Existing classification methods are designed to classify unknown instances within a set of previously known classes that are seen in training. Such classification takes the form of prediction within a closed-set. However, a more realistic scenario that fits the ground truth of real world applications is to consider the possibility of encountering instances that do not belong to any of the classes that are seen in training, $i.e.$, an open-set classification. In such situation, existing closed-set classification methods will assign a training label to these instances resulting in a misclassification. In this paper, we introduce Galaxy-X, a novel multi-class classification method for open-set problem. For each class of the training set, Galaxy-X creates a minimum bounding hyper-sphere that encompasses the distribution of the class by enclosing all of its instances. In such manner, our method is able to distinguish instances resembling previously seen classes from those that are of unseen classes. Experimental results on benchmark datasets show the efficiency of our approach in classifying novel instances from known as well as unknown classes. We also introduce a novel evaluation procedure to adequately evaluate open-set classification.

Statistical analysis of coupled time series with Kernel Cross-Spectral Density operators.

Many applications require the analysis of complex interactions between time series. These interactions can be non-linear and involve vector valued as well as complex data structures such as graphs or strings. Here we provide a general framework for the statistical analysis of these interactions when random variables are sampled from stationary time-series of arbitrary objects. To achieve this goal we analyze the properties of the kernel cross-spectral density operator induced by positive definite kernels on arbitrary input domains. This framework enables us to develop an independence test between time series as well as a similarity measure to compare different types of coupling. The performance of our test is compared to the HSIC test using i.i.d. assumptions, showing improvement in terms of detection errors as well as the suitability of this approach for testing dependency in complex dynamical systems. Finally, we use this approach to characterize complex interactions in electrophysiological neural time series.

On the Convergence and Robustness of Training GANs with Regularized Optimal Transport

Generative Adversarial Networks (GANs) are one of the most practical methods for learning data distributions. A popular GAN formulation is based on the use of Wasserstein distance as a metric between probability distributions. Unfortunately, minimizing the Wasserstein distance between the data distribution and the generative model distribution is a computationally challenging problem as its objective is non-convex, non-smooth, and even hard to compute. In this work, we show that obtaining gradient information of the smoothed Wasserstein GAN formulation, which is based on regularized Optimal Transport (OT), is computationally effortless and hence one can apply first order optimization methods to minimize this objective. Consequently, we establish theoretical convergence guarantee to stationarity for a proposed class of GAN optimization algorithms. Unlike the original non-smooth formulation, our algorithm only requires solving the discriminator to approximate optimality. We apply our method to learning MNIST digits as well as CIFAR-10 images. Our experiments show that our method is computationally efficient and generates images comparable to the state of the art algorithms given the same architecture and computational power.

An optimal algorithm for the Thresholding Bandit Problem

We study a specific \textit{combinatorial pure exploration stochastic bandit problem} where the learner aims at finding the set of arms whose means are above a given threshold, up to a given precision, and \textit{for a fixed time horizon}. We propose a parameter-free algorithm based on an original heuristic, and prove that it is optimal for this problem by deriving matching upper and lower bounds. To the best of our knowledge, this is the first non-trivial pure exploration setting with \textit{fixed budget} for which optimal strategies are constructed.

DeepPINK: reproducible feature selection in deep neural networks

Deep learning has become increasingly popular in both supervised and unsupervised machine learning thanks to its outstanding empirical performance. However, because of their intrinsic complexity, most deep learning methods are largely treated as black box tools with little interpretability. Even though recent attempts have been made to facilitate the interpretability of deep neural networks (DNNs), existing methods are susceptible to noise and lack of robustness. Therefore, scientists are justifiably cautious about the reproducibility of the discoveries, which is often related to the interpretability of the underlying statistical models. In this paper, we describe a method to increase the interpretability and reproducibility of DNNs by incorporating the idea of feature selection with controlled error rate. By designing a new DNN architecture and integrating it with the recently proposed knockoffs framework, we perform feature selection with a controlled error rate, while maintaining high power. This new method, DeepPINK (Deep feature selection using Paired-Input Nonlinear Knockoffs), is applied to both simulated and real data sets to demonstrate its empirical utility.

Mollifying Networks

The optimization of deep neural networks can be more challenging than traditional convex optimization problems due to the highly non-convex nature of the loss function, e.g. it can involve pathological landscapes such as saddle-surfaces that can be difficult to escape for algorithms based on simple gradient descent. In this paper, we attack the problem of optimization of highly non-convex neural networks by starting with a smoothed -- or \textit{mollified} -- objective function that gradually has a more non-convex energy landscape during the training. Our proposition is inspired by the recent studies in continuation methods: similar to curriculum methods, we begin learning an easier (possibly convex) objective function and let it evolve during the training, until it eventually goes back to being the original, difficult to optimize, objective function. The complexity of the mollified networks is controlled by a single hyperparameter which is annealed during the training. We show improvements on various difficult optimization tasks and establish a relationship with recent works on continuation methods for neural networks and mollifiers.

Learning What and Where to Draw

Generative Adversarial Networks (GANs) have recently demonstrated the capability to synthesize compelling real-world images, such as room interiors, album covers, manga, faces, birds, and flowers. While existing models can synthesize images based on global constraints such as a class label or caption, they do not provide control over pose or object location. We propose a new model, the Generative Adversarial What-Where Network (GAWWN), that synthesizes images given instructions describing what content to draw in which location. We show high-quality 128 × 128 image synthesis on the Caltech-UCSD Birds dataset, conditioned on both informal text descriptions and also object location. Our system exposes control over both the bounding box around the bird and its constituent parts. By modeling the conditional distributions over part locations, our system also enables conditioning on arbitrary subsets of parts (e.g. only the beak and tail), yielding an efficient interface for picking part locations.

Fashion DNA: Merging Content and Sales Data for Recommendation and Article Mapping

We present a method to determine Fashion DNA, coordinate vectors locating fashion items in an abstract space. Our approach is based on a deep neural network architecture that ingests curated article information such as tags and images, and is trained to predict sales for a large set of frequent customers. In the process, a dual space of customer style preferences naturally arises. Interpretation of the metric of these spaces is straightforward: The product of Fashion DNA and customer style vectors yields the forecast purchase likelihood for the customer-item pair, while the angle between Fashion DNA vectors is a measure of item similarity. Importantly, our models are able to generate unbiased purchase probabilities for fashion items based solely on article information, even in absence of sales data, thus circumventing the "cold-start problem" of collaborative recommendation approaches. Likewise, it generalizes easily and reliably to customers outside the training set. We experiment with Fashion DNA models based on visual and/or tag item data, evaluate their recommendation power, and discuss the resulting article similarities.

Causal Inference in the Presence of Latent Variables and Selection Bias

We show that there is a general, informative and reliable procedure for discovering causal relations when, for all the investigator knows, both latent variables and selection bias may be at work. Given information about conditional independence and dependence relations between measured variables, even when latent variables and selection bias may be present, there are sufficient conditions for reliably concluding that there is a causal path from one variable to another, and sufficient conditions for reliably concluding when no such causal path exists.

Integrated accounts of behavioral and neuroimaging data using flexible recurrent neural network models

Neuroscience studies of human decision-making abilities commonly involve subjects completing a decision-making task while BOLD signals are recorded using fMRI. Hypotheses are tested about which brain regions mediate the effect of past experience, such as rewards, on future actions. One standard approach to this is model-based fMRI data analysis, in which a model is fitted to the behavioral data, i.e., a subject's choices, and then the neural data are parsed to find brain regions whose BOLD signals are related to the model's internal signals. However, the internal mechanics of such purely behavioral models are not constrained by the neural data, and therefore might miss or mischaracterize aspects of the brain. To address this limitation, we introduce a new method using recurrent neural network models that are flexible enough to be jointly fitted to the behavioral and neural data. We trained a model so that its internal states were suitably related to neural activity during the task, while at the same time its output predicted the next action a subject would execute. We then used the fitted model to create a novel visualization of the relationship between the activity in brain regions at different times following a reward and the choices the subject subsequently made. Finally, we validated our method using a previously published dataset. We found that the model was able to recover the underlying neural substrates that were discovered by explicit model engineering in the previous work, and also derived new results regarding the temporal pattern of brain activity.

Uprooting and Rerooting Higher-Order Graphical Models

The idea of uprooting and rerooting graphical models was introduced specifically for binary pairwise models by Weller (2016) as a way to transform a model to any of a whole equivalence class of related models, such that inference on any one model yields inference results for all others. This is very helpful since inference, or relevant bounds, may be much easier to obtain or more accurate for some model in the class. Here we introduce methods to extend the approach to models with higher-order potentials and develop theoretical insights. In particular, we show that the triplet-consistent polytope TRI is unique in being `universally rooted'. We demonstrate empirically that rerooting can significantly improve accuracy of methods of inference for higher-order models at negligible computational cost.

Suppressing the Unusual: towards Robust CNNs using Symmetric Activation Functions

Many deep Convolutional Neural Networks (CNN) make incorrect predictions on adversarial samples obtained by imperceptible perturbations of clean samples. We hypothesize that this is caused by a failure to suppress unusual signals within network layers. As remedy we propose the use of Symmetric Activation Functions (SAF) in non-linear signal transducer units. These units suppress signals of exceptional magnitude. We prove that SAF networks can perform classification tasks to arbitrary precision in a simplified situation. In practice, rather than use SAFs alone, we add them into CNNs to improve their robustness. The modified CNNs can be easily trained using popular strategies with the moderate training load. Our experiments on MNIST and CIFAR-10 show that the modified CNNs perform similarly to plain ones on clean samples, and are remarkably more robust against adversarial and nonsense samples.

Ontology Enrichment by Extracting Hidden Assertional Knowledge from Text

In this position paper we present a new approach for discovering some special classes of assertional knowledge in the text by using large RDF repositories, resulting in the extraction of new non-taxonomic ontological relations. Also we use inductive reasoning beside our approach to make it outperform. Then, we prepare a case study by applying our approach on sample data and illustrate the soundness of our proposed approach. Moreover in our point of view current LOD cloud is not a suitable base for our proposal in all informational domains. Therefore we figure out some directions based on prior works to enrich datasets of Linked Data by using web mining. The result of such enrichment can be reused for further relation extraction and ontology enrichment from unstructured free text documents.

Towards Robust Detection of Adversarial Examples

Although the recent progress is substantial, deep learning methods can be vulnerable to the maliciously generated adversarial examples. In this paper, we present a novel training procedure and a thresholding test strategy, towards robust detection of adversarial examples. In training, we propose to minimize the reverse cross-entropy (RCE), which encourages a deep network to learn latent representations that better distinguish adversarial examples from normal ones. In testing, we propose to use a thresholding strategy as the detector to filter out adversarial examples for reliable predictions. Our method is simple to implement using standard algorithms, with little extra training cost compared to the common cross-entropy minimization. We apply our method to defend various attacking methods on the widely used MNIST and CIFAR-10 datasets, and achieve significant improvements on robust predictions under all the threat models in the adversarial setting.

Stochastic Ratio Matching of RBMs for Sparse High-Dimensional Inputs

Sparse high-dimensional data vectors are common in many application domains where a very large number of rarely non-zero features can be devised. Unfortunately, this creates a computational bottleneck for unsupervised feature learning algorithms such as those based on auto-encoders and RBMs, because they involve a reconstruction step where the whole input vector is predicted from the current feature values. An algorithm was recently developed to successfully handle the case of auto-encoders, based on an importance sampling scheme stochastically selecting which input elements to actually reconstruct during training for each particular example. To generalize this idea to RBMs, we propose a stochastic ratio-matching algorithm that inherits all the computational advantages and unbiasedness of the importance sampling scheme. We show that stochastic ratio matching is a good estimator, allowing the approach to beat the state-of-the-art on two bag-of-word text classification benchmarks (20 Newsgroups and RCV1), while keeping computational cost linear in the number of non-zeros.

Analyzing Games with Ambiguous Player Types using the ${\rm MINthenMAX}$ Decision Model

In many common interactive scenarios, participants lack information about other participants, and specifically about the preferences of other participants. In this work, we model an extreme case of incomplete information, which we term games with type ambiguity, where a participant lacks even information enabling him to form a belief on the preferences of others. Under type ambiguity, one cannot analyze the scenario using the commonly used Bayesian framework, and therefore he needs to model the participants using a different decision model.

On the Generalization Ability of Online Learning Algorithms for Pairwise Loss Functions

In this paper, we study the generalization properties of online learning based stochastic methods for supervised learning problems where the loss function is dependent on more than one training sample (e.g., metric learning, ranking). We present a generic decoupling technique that enables us to provide Rademacher complexity-based generalization error bounds. Our bounds are in general tighter than those obtained by Wang et al (COLT 2012) for the same problem. Using our decoupling technique, we are further able to obtain fast convergence rates for strongly convex pairwise loss functions. We are also able to analyze a class of memory efficient online learning algorithms for pairwise learning problems that use only a bounded subset of past training samples to update the hypothesis at each step. Finally, in order to complement our generalization bounds, we propose a novel memory efficient online learning algorithm for higher order learning problems with bounded regret guarantees.

Learning Dexterous Manipulation Policies from Experience and Imitation

We explore learning-based approaches for feedback control of a dexterous five-finger hand performing non-prehensile manipulation. First, we learn local controllers that are able to perform the task starting at a predefined initial state. These controllers are constructed using trajectory optimization with respect to locally-linear time-varying models learned directly from sensor data. In some cases, we initialize the optimizer with human demonstrations collected via teleoperation in a virtual environment. We demonstrate that such controllers can perform the task robustly, both in simulation and on the physical platform, for a limited range of initial conditions around the trained starting state. We then consider two interpolation methods for generalizing to a wider range of initial conditions: deep learning, and nearest neighbors. We find that nearest neighbors achieve higher performance. Nevertheless, the neural network has its advantages: it uses only tactile and proprioceptive feedback but no visual feedback about the object (i.e. it performs the task blind) and learns a time-invariant policy. In contrast, the nearest neighbors method switches between time-varying local controllers based on the proximity of initial object states sensed via motion capture. While both generalization methods leave room for improvement, our work shows that (i) local trajectory-based controllers for complex non-prehensile manipulation tasks can be constructed from surprisingly small amounts of training data, and (ii) collections of such controllers can be interpolated to form more global controllers. Results are summarized in the supplementary video:

Trained Ternary Quantization

Deep neural networks are widely used in machine learning applications. However, the deployment of large neural networks models can be difficult to deploy on mobile devices with limited power budgets. To solve this problem, we propose Trained Ternary Quantization (TTQ), a method that can reduce the precision of weights in neural networks to ternary values. This method has very little accuracy degradation and can even improve the accuracy of some models (32, 44, 56-layer ResNet) on CIFAR-10 and AlexNet on ImageNet. And our AlexNet model is trained from scratch, which means it’s as easy as to train normal full precision model. We highlight our trained quantization method that can learn both ternary values and ternary assignment. During inference, only ternary values (2-bit weights) and scaling factors are needed, therefore our models are nearly 16× smaller than full- precision models. Our ternary models can also be viewed as sparse binary weight networks, which can potentially be accelerated with custom circuit. Experiments on CIFAR-10 show that the ternary models obtained by trained quantization method outperform full-precision models of ResNet-32,44,56 by 0.04%, 0.16%, 0.36%, respectively. On ImageNet, our model outperforms full-precision AlexNet model by 0.3% of Top-1 accuracy and outperforms previous ternary models by 3%.

Fine-grained Optimization of Deep Neural Networks

To this end, we pose two problems. First, we aim to obtain weights whose different norms are all upper bounded by a constant number. To achieve these bounds, we propose a two-stage renormalization procedure; (i) normalization of weights according to different norms used in the bounds, and (ii) reparameterization of the normalized weights to set a constant and finite upper bound of their norms. In the second problem, we consider training DNNs with these renormalized weights. To this end, we first propose a strategy to construct joint spaces (manifolds) of weights according to different constraints in DNNs. Next, we propose a fine-grained SGD algorithm (FG-SGD) for optimization on the weight manifolds to train DNNs with assurance of convergence to minima. Experimental analyses show that image classification accuracy of baseline DNNs can be boosted using FG-SGD on collections of manifolds identified by multiple constraints.

Texture Classification Approach Based on Combination of Edge & Co-occurrence and Local Binary Pattern

Texture classification is one of the problems which has been paid much attention on by computer scientists since late 90s. If texture classification is done correctly and accurately, it can be used in many cases such as Pattern recognition, object tracking, and shape recognition. So far, there have been so many methods offered to solve this problem. Near all these methods have tried to extract and define features to separate different labels of textures really well. This article has offered an approach which has an overall process on the images of textures based on Local binary pattern and Gray Level Co-occurrence matrix and then by edge detection, and finally, extracting the statistical features from the images would classify them. Although, this approach is a general one and is could be used in different applications, the method has been tested on the stone texture and the results have been compared with some of the previous approaches to prove the quality of proposed approach.

The Computational Complexity of Sensitivity Analysis and Parameter Tuning

While known algorithms for sensitivity analysis and parameter tuning in probabilistic networks have a running time that is exponential in the size of the network, the exact computational complexity of these problems has not been established as yet. In this paper we study several variants of the tuning problem and show that these problems are NPPP-complete in general. We further show that the problems remain NP-complete or PP-complete, for a number of restricted variants. These complexity results provide insight in whether or not recent achievements in sensitivity analysis and tuning can be extended to more general, practicable methods.