Feature learning forms the cornerstone for tackling challenging learning problems in domains such as speech, computer vision and natural language processing. In this paper, we consider a novel class of matrix and tensor-valued features, which can be pre-trained using unlabeled samples. We present efficient algorithms for extracting discriminative information, given these pre-trained features and labeled samples for any related task. Our class of features are based on higher-order score functions, which capture local variations in the probability density function of the input. We establish a theoretical framework to characterize the nature of discriminative information that can be extracted from score-function features, when used in conjunction with labeled samples. We employ efficient spectral decomposition algorithms (on matrices and tensors) for extracting discriminative components. The advantage of employing tensor-valued features is that we can extract richer discriminative information in the form of an overcomplete representations. Thus, we present a novel framework for employing generative models of the input for discriminative learning.

Multithreshold Entropy Linear Classifier (MELC) is a recent classifier idea which employs information theoretic concept in order to create a multithreshold maximum margin model. In this paper we analyze its consistency over multithreshold linear models and show that its objective function upper bounds the amount of misclassified points in a similar manner like hinge loss does in support vector machines. For further confirmation we also conduct some numerical experiments on five datasets.

Multithreshold Entropy Linear Classifier (MELC) is a density based model which searches for a linear projection maximizing the Cauchy-Schwarz Divergence of dataset kernel density estimation. Despite its good empirical results, one of its drawbacks is the optimization speed. In this paper we analyze how one can speed it up through solving an approximate problem. We analyze two methods, both similar to the approximate solutions of the Kernel Density Estimation querying and provide adaptive schemes for selecting a crucial parameters based on user-specified acceptable error. Furthermore we show how one can exploit well known conjugate gradients and L-BFGS optimizers despite the fact that the original optimization problem should be solved on the sphere. All above methods and modifications are tested on 10 real life datasets from UCI repository to confirm their practical usability.

In this paper, we present GASG21 (Grassmannian Adaptive Stochastic Gradient for $L_{2,1}$ norm minimization), an adaptive stochastic gradient algorithm to robustly recover the low-rank subspace from a large matrix. In the presence of column outliers, we reformulate the batch mode matrix $L_{2,1}$ norm minimization with rank constraint problem as a stochastic optimization approach constrained on Grassmann manifold. For each observed data vector, the low-rank subspace $\mathcal{S}$ is updated by taking a gradient step along the geodesic of Grassmannian. In order to accelerate the convergence rate of the stochastic gradient method, we choose to adaptively tune the constant step-size by leveraging the consecutive gradients. Furthermore, we demonstrate that with proper initialization, the K-subspaces extension, K-GASG21, can robustly cluster a large number of corrupted data vectors into a union of subspaces. Numerical experiments on synthetic and real data demonstrate the efficiency and accuracy of the proposed algorithms even with heavy column outliers corruption.

Modality is one of the important components of grammar in linguistics. It lets speaker to express attitude towards, or give assessment or potentiality of state of affairs. It implies different senses and thus has different perceptions as per the context. This paper presents an account showing the gap in the functionality of the current state of art Natural Language Processing (NLP) systems. The contextual nature of linguistic modality is studied. In this paper, the works and logical approaches employed by Natural Language Processing systems dealing with modality are reviewed. It sees human cognition and intelligence as multi-layered approach that can be implemented by intelligent systems for learning. Lastly, current flow of research going on within this field is talked providing futurology.

The performance of trust models highly depend on the characteristics of the environments where they are applied. Thus, it becomes challenging to choose a suitable trust model for a given e-marketplace environment, especially when ground truth about the agent (buyer and seller) behavior is unknown (called unknown environment). We propose a case-based reasoning framework to choose suitable trust models for unknown environments, based on the intuition that if a trust model performs well in one environment, it will do so in another similar environment. Firstly, we build a case base with a number of simulated environments (with known ground truth) along with the trust models most suitable for each of them. Given an unknown environment, case-based retrieval algorithms retrieve the most similar case(s), and the trust model of the most similar case(s) is chosen as the most suitable model for the unknown environment. Evaluation results confirm the effectiveness of our framework in choosing suitable trust models for different e-marketplace environments.

We consider the problem of closeness testing for two discrete distributions in the practically relevant setting of \emph{unequal} sized samples drawn from each of them. Specifically, given a target error parameter $\varepsilon > 0$, $m_1$ independent draws from an unknown distribution $p,$ and $m_2$ draws from an unknown distribution $q$, we describe a test for distinguishing the case that $p=q$ from the case that $||p-q||_1 \geq \varepsilon$. If $p$ and $q$ are supported on at most $n$ elements, then our test is successful with high probability provided $m_1\geq n^{2/3}/\varepsilon^{4/3}$ and $m_2 = \Omega(\max\{\frac{n}{\sqrt m_1\varepsilon^2}, \frac{\sqrt n}{\varepsilon^2}\});$ we show that this tradeoff is optimal throughout this range, to constant factors. These results extend the recent work of Chan et al. who established the sample complexity when the two samples have equal sizes, and tightens the results of Acharya et al. by polynomials factors in both $n$ and $\varepsilon$. As a consequence, we obtain an algorithm for estimating the mixing time of a Markov chain on $n$ states up to a $\log n$ factor that uses $\tilde{O}(n^{3/2} \tau_{mix})$ queries to a "next node" oracle, improving upon the $\tilde{O}(n^{5/3}\tau_{mix})$ query algorithm of Batu et al. Finally, we note that the core of our testing algorithm is a relatively simple statistic that seems to perform well in practice, both on synthetic data and on natural language data.

In data science and machine learning, hierarchical parametric models, such as mixture models, are often used. They contain two kinds of variables: observable variables, which represent the parts of the data that can be directly measured, and latent variables, which represent the underlying processes that generate the data. Although there has been an increase in research on the estimation accuracy for observable variables, the theoretical analysis of estimating latent variables has not been thoroughly investigated. In a previous study, we determined the accuracy of a Bayes estimation for the joint probability of the latent variables in a dataset, and we proved that the Bayes method is asymptotically more accurate than the maximum-likelihood method. However, the accuracy of the Bayes estimation for a single latent variable remains unknown. In the present paper, we derive the asymptotic expansions of the error functions, which are defined by the Kullback-Leibler divergence, for two types of single-variable estimations when the statistical regularity is satisfied. Our results indicate that the accuracies of the Bayes and maximum-likelihood methods are asymptotically equivalent and clarify that the Bayes method is only advantageous for multivariable estimations.

We are proposing an extension of the recursive neural network that makes use of a variant of the long short-term memory architecture. The extension allows information low in parse trees to be stored in a memory register (the `memory cell') and used much later higher up in the parse tree. This provides a solution to the vanishing gradient problem and allows the network to capture long range dependencies. Experimental results show that our composition outperformed the traditional neural-network composition on the Stanford Sentiment Treebank.

Apple is refusing to explain how and why the Russian language version of its voice-controlled virtual assistant, Siri, provided homophobic answers to queries relating to gay or lesbian topics. The allegations first came to light when a Russian man called Alex, who lives in London, uploaded a video to YouTube appearing to show Siri either evading questions or expressing a negative response. Alex asked questions such as: "Are there any gay bars around me?", "tell me about gay marriage?" "I would have turned red if I could", "you are so rude!" and "I will pretend I haven't heard". These responses seem to suggest that the word gay () in Russian has been programmed as profanity. Swear words in the English language version receive similar responses.