Taking into account high-order interactions among covariates is valuable in many practical regression problems. This is, however, computationally challenging task because the number of high-order interaction features to be considered would be extremely large unless the number of covariates is sufficiently small. In this paper, we propose a novel efficient algorithm for LASSO-based sparse learning of such high-order interaction models. Our basic strategy for reducing the number of features is to employ the idea of recently proposed safe feature screening (SFS) rule. An SFS rule has a property that, if a feature satisfies the rule, then the feature is guaranteed to be non-active in the LASSO solution, meaning that it can be safely screened-out prior to the LASSO training process. If a large number of features can be screened-out before training the LASSO, the computational cost and the memory requirment can be dramatically reduced. However, applying such an SFS rule to each of the extremely large number of high-order interaction features would be computationally infeasible. Our key idea for solving this computational issue is to exploit the underlying tree structure among high-order interaction features. Specifically, we introduce a pruning condition called safe feature pruning (SFP) rule which has a property that, if the rule is satisfied in a certain node of the tree, then all the high-order interaction features corresponding to its descendant nodes can be guaranteed to be non-active at the optimal solution. Our algorithm is extremely efficient, making it possible to work, e.g., with 3rd order interactions of 10,000 original covariates, where the number of possible high-order interaction features is greater than 10^{12}.

Finding statistically significant high-order interaction features in predictive modeling is important but challenging task. The difficulty lies in the fact that, for a recent applications with high-dimensional covariates, the number of possible high-order interaction features would be extremely large. Identifying statistically significant features from such a huge pool of candidates would be highly challenging both in computational and statistical senses. To work with this problem, we consider a two stage algorithm where we first select a set of high-order interaction features by marginal screening, and then make statistical inferences on the regression model fitted only with the selected features. Such statistical inferences are called post-selection inference (PSI), and receiving an increasing attention in the literature. One of the seminal recent advancements in PSI literature is the works by Lee et al. where the authors presented an algorithmic framework for computing exact sampling distributions in PSI. A main challenge when applying their approach to our high-order interaction models is to cope with the fact that PSI in general depends not only on the selected features but also on the unselected features, making it hard to apply to our extremely high-dimensional high-order interaction models. The goal of this paper is to overcome this difficulty by introducing a novel efficient method for PSI. Our key idea is to exploit the underlying tree structure among high-order interaction features, and to develop a pruning method of the tree which enables us to quickly identify a group of unselected features that are guaranteed to have no influence on PSI. The experimental results indicate that the proposed method allows us to reliably identify statistically significant high-order interaction features with reasonable computational cost.

The New Yorker publishes a weekly captionless cartoon. More than 5,000 readers submit captions for it. The editors select three of them and ask the readers to pick the funniest one. We describe an experiment that compares a dozen automatic methods for selecting the funniest caption. We show that negative sentiment, human-centeredness, and lexical centrality most strongly match the funniest captions, followed by positive sentiment. These results are useful for understanding humor and also in the design of more engaging conversational agents in text and multimodal (vision+text) systems. As part of this work, a large set of cartoons and captions is being made available to the community.

Stochastic variational inference (SVI) is emerging as the most promising candidate for scaling inference in Bayesian probabilistic models to large datasets. However, the performance of these methods has been assessed primarily in the context of Bayesian topic models, particularly latent Dirichlet allocation (LDA). Deriving several new algorithms, and using synthetic, image and genomic datasets, we investigate whether the understanding gleaned from LDA applies in the setting of sparse latent factor models, specifically beta process factor analysis (BPFA). We demonstrate that the big picture is consistent: using Gibbs sampling within SVI to maintain certain posterior dependencies is extremely effective. However, we find that different posterior dependencies are important in BPFA relative to LDA. Particularly, approximations able to model intra-local variable dependence perform best.

The modelling of data on a spherical surface requires the consideration of directional probability distributions. To model asymmetrically distributed data on a three-dimensional sphere, Kent distributions are often used. The moment estimates of the parameters are typically used in modelling tasks involving Kent distributions. However, these lack a rigorous statistical treatment. The focus of the paper is to introduce a Bayesian estimation of the parameters of the Kent distribution which has not been carried out in the literature, partly because of its complex mathematical form. We employ the Bayesian information-theoretic paradigm of Minimum Message Length (MML) to bridge this gap and derive reliable estimators. The inferred parameters are subsequently used in mixture modelling of Kent distributions. The problem of inferring the suitable number of mixture components is also addressed using the MML criterion. We demonstrate the superior performance of the derived MML-based parameter estimates against the traditional estimators. We apply the MML principle to infer mixtures of Kent distributions to model empirical data corresponding to protein conformations. We demonstrate the effectiveness of Kent models to act as improved descriptors of protein structural data as compared to commonly used von Mises-Fisher distributions.

Factorial hidden Markov models (FHMMs) are powerful tools of modeling sequential data. Learning FHMMs yields a challenging simultaneous model selection issue, i.e., selecting the number of multiple Markov chains and the dimensionality of each chain. Our main contribution is to address this model selection issue by extending Factorized Asymptotic Bayesian (FAB) inference to FHMMs. First, we offer a better approximation of marginal log-likelihood than the previous FAB inference. Our key idea is to integrate out transition probabilities, yet still apply the Laplace approximation to emission probabilities. Second, we prove that if there are two very similar hidden states in an FHMM, i.e. one is redundant, then FAB will almost surely shrink and eliminate one of them, making the model parsimonious. Experimental results show that FAB for FHMMs significantly outperforms state-of-the-art nonparametric Bayesian iFHMM and Variational FHMM in model selection accuracy, with competitive held-out perplexity.

This work studies two interrelated problems - online robust PCA (RPCA) and online low-rank matrix completion (MC). In recent work by Cand\`{e}s et al., RPCA has been defined as a problem of separating a low-rank matrix (true data), $L:=[\ell_1, \ell_2, \dots \ell_{t}, \dots , \ell_{t_{\max}}]$ and a sparse matrix (outliers), $S:=[x_1, x_2, \dots x_{t}, \dots, x_{t_{\max}}]$ from their sum, $M:=L+S$. Our work uses this definition of RPCA. An important application where both these problems occur is in video analytics in trying to separate sparse foregrounds (e.g., moving objects) and slowly changing backgrounds. While there has been a large amount of recent work on both developing and analyzing batch RPCA and batch MC algorithms, the online problem is largely open. In this work, we develop a practical modification of our recently proposed algorithm to solve both the online RPCA and online MC problems. The main contribution of this work is that we obtain correctness results for the proposed algorithms under mild assumptions. The assumptions that we need are: (a) a good estimate of the initial subspace is available (easy to obtain using a short sequence of background-only frames in video surveillance); (b) the $\ell_t$'s obey a `slow subspace change' assumption; (c) the basis vectors for the subspace from which $\ell_t$ is generated are dense (non-sparse); (d) the support of $x_t$ changes by at least a certain amount at least every so often; and (e) algorithm parameters are appropriately set

Canonical Correlation Analysis (CCA) is a widely used spectral technique for finding correlation structures in multi-view datasets. In this paper, we tackle the problem of large scale CCA, where classical algorithms, usually requiring computing the product of two huge matrices and huge matrix decomposition, are computationally and storage expensive. We recast CCA from a novel perspective and propose a scalable and memory efficient Augmented Approximate Gradient (AppGrad) scheme for finding top $k$ dimensional canonical subspace which only involves large matrix multiplying a thin matrix of width $k$ and small matrix decomposition of dimension $k\times k$. Further, AppGrad achieves optimal storage complexity $O(k(p_1+p_2))$, compared with classical algorithms which usually require $O(p_1^2+p_2^2)$ space to store two dense whitening matrices. The proposed scheme naturally generalizes to stochastic optimization regime, especially efficient for huge datasets where batch algorithms are prohibitive. The online property of stochastic AppGrad is also well suited to the streaming scenario, where data comes sequentially. To the best of our knowledge, it is the first stochastic algorithm for CCA. Experiments on four real data sets are provided to show the effectiveness of the proposed methods.

Fairness-aware learning is a novel framework for classification tasks. Like regular empirical risk minimization (ERM), it aims to learn a classifier with a low error rate, and at the same time, for the predictions of the classifier to be independent of sensitive features, such as gender, religion, race, and ethnicity. Existing methods can achieve low dependencies on given samples, but this is not guaranteed on unseen samples. The existing fairness-aware learning algorithms employ different dependency measures, and each algorithm is specifically designed for a particular one. Such diversity makes it difficult to theoretically analyze and compare them. In this paper, we propose a general framework for fairness-aware learning that uses f-divergences and that covers most of the dependency measures employed in the existing methods. We introduce a way to estimate the f-divergences that allows us to give a unified analysis for the upper bound of the estimation error; this bound is tighter than that of the existing convergence rate analysis of the divergence estimation. With our divergence estimate, we propose a fairness-aware learning algorithm, and perform a theoretical analysis of its generalization error. Our analysis reveals that, under mild assumptions and even with enforcement of fairness, the generalization error of our method is $O(\sqrt{1/n})$, which is the same as that of the regular ERM. In addition, and more importantly, we show that, for any f-divergence, the upper bound of the estimation error of the divergence is $O(\sqrt{1/n})$. This indicates that our fairness-aware learning algorithm guarantees low dependencies on unseen samples for any dependency measure represented by an f-divergence.

The recent explosion in the amount and dimensionality of data has exacerbated the need of trading off computational and statistical efficiency carefully, so that inference is both tractable and meaningful. We propose a framework that provides an explicit opportunity for practitioners to specify how much statistical risk they are willing to accept for a given computational cost, and leads to a theoretical risk-computation frontier for any given inference problem. We illustrate the tradeoff between risk and computation and illustrate the frontier in three distinct settings. First, we derive analytic forms for the risk of estimating parameters in the classical setting of estimating the mean and variance for normally distributed data and for the more general setting of parameters of an exponential family. The second example concentrates on computationally constrained Hodges-Lehmann estimators. We conclude with an evaluation of risk associated with early termination of iterative matrix inversion algorithms in the context of linear regression.