The marginal log-likelihood is a key concept of Bayesian model identification of latent variable models (LVMs), such as mixture models (MMs), probabilistic principal component analysis, and hidden Markov models (HMMs). Determination of dimensionality of latent variables is an essential task to uncover hidden structures behind the observed data as well as to mitigate overfitting. In general, LVMs are singular (i.e., mapping between parameters and probabilistic models is not one-to-one) and such classical information criteria based on the regularity assumption as the Bayesian information criterion (BIC) [Schwarz, 1978] are no longer justified. Since exact evaluation of 1 the marginal log-likelihood is often not available, approximation techniques have been developed using sampling (i.e., Markov Chain Monte Carlo methods (MCMCs) [Hastings, 1970]), a variational lower bound (i.e., the variational Bayes methods (VB) [Attias, 1999, Jordan et al., 1999]), or algebraic geometry (i.e., the widely applicable BIC (WBIC) [Watanabe, 2013]). However, model selection using these methods typically requires heavy computational cost (e.g., a large number of MCMC sampling in a high-dimensional space, an outer loop for VB/WBIC.) In the last few years, a new approximation technique and an inference method, factorized information criterion (FIC) and factorized asymptotic Bayesian inference (FAB), have been developed for some binary LVMs [Fujimaki and Morinaga, 2012, Fujimaki and Hayashi, 2012, Hayashi and Fujimaki, 2013, Eto et al., 2014]. Unlike existing methods which evaluate approximated marginal log-likelihoods calculated for each latent variable dimensionality (and therefore need an outer loop for model selection), FAB finds an effective dimensionality via an EMstyle alternating optimization procedure.

Community detection in graphs has been extensively studied both in theory and in applications. However, detecting communities in hypergraphs is more challenging. In this paper, we propose a tensor decomposition approach for guaranteed learning of communities in a special class of hypergraphs modeling social tagging systems or folksonomies. A folksonomy is a tripartite 3-uniform hypergraph consisting of (user, tag, resource) hyperedges. We posit a probabilistic mixed membership community model, and prove that the tensor method consistently learns the communities under efficient sample complexity and separation requirements.

For random graphs distributed according to stochastic blockmodels, a special case of latent position graphs, adjacency spectral embedding followed by appropriate vertex classification is asymptotically Bayes optimal; but this approach requires knowledge of and critically depends on the model dimension. In this paper, we propose a sparse representation vertex classifier which does not require information about the model dimension. This classifier represents a test vertex as a sparse combination of the vertices in the training set and uses the recovered coefficients to classify the test vertex. We prove consistency of our proposed classifier for stochastic blockmodels, and demonstrate that the sparse representation classifier can predict vertex labels with higher accuracy than adjacency spectral embedding approaches via both simulation studies and real data experiments. Our results demonstrate the robustness and effectiveness of our proposed vertex classifier when the model dimension is unknown.

A wide variety of problems in machine learning, including exemplar clustering, document summarization, and sensor placement, can be cast as constrained submodular maximization problems. Unfortunately, the resulting submodular optimization problems are often too large to be solved on a single machine. We develop a simple distributed algorithm that is embarrassingly parallel and it achieves provable, constant factor, worst-case approximation guarantees. In our experiments, we demonstrate its efficiency in large problems with different kinds of constraints with objective values always close to what is achievable in the centralized setting.

Every day, billions of mobile network events (i.e. CDRs) are generated by cellular phone operator companies. Latent in this data are inspiring insights about human actions and behaviors, the discovery of which is important because context-aware applications and services hold the key to user-driven, intelligent services, which can enhance our everyday lives such as social and economic development, urban planning, and health prevention. The major challenge in this area is that interpreting such a big stream of data requires a deep understanding of mobile network events' context through available background knowledge. This article addresses the issues in context awareness given heterogeneous and uncertain data of mobile network events missing reliable information on the context of this activity. The contribution of this research is a model from a combination of logical and statistical reasoning standpoints for enabling human activity inference in qualitative terms from open geographical data that aimed at improving the quality of human behaviors recognition tasks from CDRs. We use open geographical data, Openstreetmap (OSM), as a proxy for predicting the content of human activity in the area. The user study performed in Trento shows that predicted human activities (top level) match the survey data with around 93% overall accuracy. The extensive validation for predicting a more specific economic type of human activity performed in Barcelona, by employing credit card transaction data. The analysis identifies that appropriately normalized data on points of interest (POI) is a good proxy for predicting human economical activities, with 84% accuracy on average. So the model is proven to be efficient for predicting the context of human activity, when its total level could be efficiently observed from cell phone data records, missing contextual information however.

In the presence of weak overall correlation, it may be useful to investigate if the correlation is significantly and substantially more pronounced over a subpopulation. Two different testing procedures are compared. Both are based on the rankings of the values of two variables from a data set with a large number n of observations. The first maintains its level against Gaussian copulas; the second adapts to general alternatives in the sense that that the number of parameters used in the test grows with n . An analysis of wine quality illustrates how the methods detect heterogeneity of association between chemical properties of the wine, which are attributable to a mix of different cultivars.

Why does Deep Learning work? What representations does it capture? How do higher-order representations emerge? We study these questions from the perspective of group theory, thereby opening a new approach towards a theory of Deep learning. One factor behind the recent resurgence of the subject is a key algorithmic step called {\em pretraining}: first search for a good generative model for the input samples, and repeat the process one layer at a time. We show deeper implications of this simple principle, by establishing a connection with the interplay of orbits and stabilizers of group actions. Although the neural networks themselves may not form groups, we show the existence of {\em shadow} groups whose elements serve as close approximations. Over the shadow groups, the pre-training step, originally introduced as a mechanism to better initialize a network, becomes equivalent to a search for features with minimal orbits. Intuitively, these features are in a way the {\em simplest}. Which explains why a deep learning network learns simple features first. Next, we show how the same principle, when repeated in the deeper layers, can capture higher order representations, and why representation complexity increases as the layers get deeper.

The paper briefly introduces multiple classifier systems and describes a new algorithm, which improves classification accuracy by means of recommendation of a proper algorithm to an object classification. This recommendation is done assuming that a classifier is likely to predict the label of the object correctly if it has correctly classified its neighbors. The process of assigning a classifier to each object is based on Formal Concept Analysis. We explain the idea of the algorithm with a toy example and describe our first experiments with real-world datasets.

We consider two connected aspects of maximum likelihood estimation of the parameter for high-dimensional discrete graphical models: the existence of the maximum likelihood estimate (mle) and its computation. When the data is sparse, there are many zeros in the contingency table and the maximum likelihood estimate of the parameter may not exist. Fienberg and Rinaldo (2012) have shown that the mle does not exists iff the data vector belongs to a face of the so-called marginal cone spanned by the rows of the design matrix of the model. Identifying these faces in high-dimension is challenging. In this paper, we take a local approach : we show that one such face, albeit possibly not the smallest one, can be identified by looking at a collection of marginal graphical models generated by induced subgraphs $G_i,i=1,\ldots,k$ of $G$. This is our first contribution. Our second contribution concerns the composite maximum likelihood estimate. When the dimension of the problem is large, estimating the parameters of a given graphical model through maximum likelihood is challenging, if not impossible. The traditional approach to this problem has been local with the use of composite likelihood based on local conditional likelihoods. A more recent development is to have the components of the composite likelihood be marginal likelihoods centred around each $v$. We first show that the estimates obtained by consensus through local conditional and marginal likelihoods are identical. We then study the asymptotic properties of the composite maximum likelihood estimate when both the dimension of the model and the sample size $N$ go to infinity.

Artificial neural networks typically have a fixed, non-linear activation function at each neuron. We have designed a novel form of piecewise linear activation function that is learned independently for each neuron using gradient descent. With this adaptive activation function, we are able to improve upon deep neural network architectures composed of static rectified linear units, achieving state-of-the-art performance on CIFAR-10 (7.51%), CIFAR-100 (30.83%), and a benchmark from high-energy physics involving Higgs boson decay modes.