When a transition probability matrix is represented as the product of two stochastic matrices, one can swap the factors of the multiplication to obtain another transition matrix that retains some fundamental characteristics of the original. Since the derived matrix can be much smaller than its precursor, this property can be exploited to create a compact version of a Markov decision process (MDP), and hence to reduce the computational cost of dynamic programming. Building on this idea, this paper presents an approximate policy iteration algorithm called policy iteration based on stochastic factorization, or PISF for short. In terms of computational complexity, PISF replaces standard policy iteration's cubic dependence on the size of the MDP with a function that grows only linearly with the number of states in the model. The proposed algorithm also enjoys nice theoretical properties: it always terminates after a finite number of iterations and returns a decision policy whose performance only depends on the quality of the stochastic factorization. In particular, if the approximation error in the factorization is sufficiently small, PISF computes the optimal value function of the MDP. The paper also discusses practical ways of factoring an MDP and illustrates the usefulness of the proposed algorithm with an application involving a large-scale decision problem of real economical interest.

Sparse high dimensional graphical model selection is a topic of much interest in modern day statistics. A popular approach is to apply l1-penalties to either (1) parametric likelihoods, or, (2) regularized regression/pseudo-likelihoods, with the latter having the distinct advantage that they do not explicitly assume Gaussianity. As none of the popular methods proposed for solving pseudo-likelihood based objective functions have provable convergence guarantees, it is not clear if corresponding estimators exist or are even computable, or if they actually yield correct partial correlation graphs. This paper proposes a new pseudo-likelihood based graphical model selection method that aims to overcome some of the shortcomings of current methods, but at the same time retain all their respective strengths. In particular, we introduce a novel framework that leads to a convex formulation of the partial covariance regression graph problem, resulting in an objective function comprised of quadratic forms. The objective is then optimized via a coordinate-wise approach. The specific functional form of the objective function facilitates rigorous convergence analysis leading to convergence guarantees; an important property that cannot be established using standard results, when the dimension is larger than the sample size, as is often the case in high dimensional applications. These convergence guarantees ensure that estimators are well-defined under very general conditions, and are always computable. In addition, the approach yields estimators that have good large sample properties and also respect symmetry. Furthermore, application to simulated/real data, timing comparisons and numerical convergence is demonstrated. We also present a novel unifying framework that places all graphical pseudo-likelihood methods as special cases of a more general formulation, leading to important insights.

We introduce novel results for approximate inference on planar graphical models using the loop calculus framework. The loop calculus (Chertkov and Chernyak, 2006b) allows to express the exact partition function Z of a graphical model as a finite sum of terms that can be evaluated once the belief propagation (BP) solution is known. In general, full summation over all correction terms is intractable. We develop an algorithm for the approach presented in Chertkov et al. (2008) which represents an efficient truncation scheme on planar graphs and a new representation of the series in terms of Pfaffians of matrices. We analyze in detail both the loop series and the Pfaffian series for models with binary variables and pairwise interactions, and show that the first term of the Pfaffian series can provide very accurate approximations. The algorithm outperforms previous truncation schemes of the loop series and is competitive with other state-of-the-art methods for approximate inference.

We introduce a means of automating machine learning (ML) for big data tasks, by performing scalable stochastic Bayesian optimisation of ML algorithm parameters and hyper-parameters. More often than not, the critical tuning of ML algorithm parameters has relied on domain expertise from experts, along with laborious hand-tuning, brute search or lengthy sampling runs. Against this background, Bayesian optimisation is finding increasing use in automating parameter tuning, making ML algorithms accessible even to non-experts. However, the state of the art in Bayesian optimisation is incapable of scaling to the large number of evaluations of algorithm performance required to fit realistic models to complex, big data. We here describe a stochastic, sparse, Bayesian optimisation strategy to solve this problem, using many thousands of noisy evaluations of algorithm performance on subsets of data in order to effectively train algorithms for big data. We provide a comprehensive benchmarking of possible sparsification strategies for Bayesian optimisation, concluding that a Nystrom approximation offers the best scaling and performance for real tasks. Our proposed algorithm demonstrates substantial improvement over the state of the art in tuning the parameters of a Gaussian Process time series prediction task on real, big data.

In this research-in-progress paper we present a new real world domain for studying the aggregation of different opinions: early stage architectural design of buildings. This is an important real world application, not only because building design and construction is one of the world's largest industries measured by global expenditures, but also because the early stage design decision making has a significant impact on the energy consumption of buildings. We present a mapping between the domain of architecture and engineering research and that of the agent models present in the literature. We study the importance of forming diverse teams when aggregating the opinions of different agents for architectural design, and also the effect of having agents optimizing for different factors of a multi-objective optimization design problem. We find that a diverse team of agents is able to provide a higher number of top ranked solutions for the early stage designer to choose from. Finally, we present the next steps for a deeper exploration of our questions.

We have proposed a novel data mining method called co-occurring cluster mining (CCM) for mining patterns from a sequence of multidimensional event data. The CCM first generates cluster candidates and then test the candidates based on clustering in the data space as well as co-occurrence degree in the event sequence. In searching appropriate clusters associated with co-occurrence, the search space is reduced by obtaining a dendrogram from a hierarchical clustering as the clustering procedure. In this paper, we show the potential of CCM with following two applications: (1) damage patterns in fuel cell and (2) earthquake occurrence patterns. In the fuel cell application, given a sequence of acoustic emission events, which comprise of waveform signal data of damages to a fuel cell, the mechanical interactions between components of the fuel cell are inferred from the mined co-occurrence patterns. Similarly, in the application of earthquakes, the interactions between distant earthquakes are extracted as co-occurrence patterns from a hypocenter catalog.

Sequential or online dimensional reduction is of interests due to the explosion of streaming data based applications and the requirement of adaptive statistical modeling, in many emerging fields, such as the modeling of energy end-use profile. Principal Component Analysis (PCA), is the classical way of dimensional reduction. However, traditional Singular Value Decomposition (SVD) based PCA fails to model data which largely deviates from Gaussian distribution. The Bregman Divergence was recently introduced to achieve a generalized PCA framework. If the random variable under dimensional reduction follows Bernoulli distribution, which occurs in many emerging fields, the generalized PCA is called Logistic PCA (LPCA). In this paper, we extend the batch LPCA to a sequential version (i.e. SLPCA), based on the sequential convex optimization theory. The convergence property of this algorithm is discussed compared to the batch version of LPCA (i.e. BLPCA), as well as its performance in reducing the dimension for multivariate binary-state systems. Its application in building energy end-use profile modeling is also investigated.

Social influence has been widely accepted to explain people's cascade behaviors and further utilized in many related applications. However, few of existing work studied the direct, microscopic and temporal impact of social influence on people's behaviors in detail. In this paper we concentrate on the behavior modeling and systematically formulate the family of behavior propagation models (BPMs) including the static models (BP and IBP), and their discrete temporal variants (DBP and DIBP). To address the temporal dynamics of behavior propagation over continuous time, we propose a continuous temporal interest-aware behavior propagation model, called CIBP. As a new member of the BPM family, CIBP exploits the continuous-temporal functions (CTFs) to model the fully-continuous dynamic variance of social influence over time. Experiments on real-world datasets evaluated the family of BPMs and demonstrated the effectiveness of our proposed approach.

Researchers have introduced the Dynamic Distributed Constraint Optimization Problem (Dynamic DCOP) formulation to model dynamically changing multi-agent coordination problems, where a dynamic DCOP is a sequence of (static canonical) DCOPs, each partially different from the DCOP preceding it. Existing work typically assumes that the problem in each time step is decoupled from the problems in other time steps, which might not hold in some applications. Therefore, in this paper, we make the following contributions: (i) We introduce a new model, called Markovian Dynamic DCOPs (MD-DCOPs), where the DCOP in the next time step is a function of the value assignments in the current time step; (ii) We introduce two distributed reinforcement learning algorithms, the Distributed RVI Q-learning algorithm and the Distributed R-learning algorithm, that balance exploration and exploitation to solve MD-DCOPs in an online manner; and (iii) We empirically evaluate them against an existing multi-arm bandit DCOP algorithm on dynamic DCOPs.

We consider the class of convex minimization problems, composed of a self-concordant function, such as the logdet metric, a convex data fidelity term h(.) and, a regularizing — possibly non-smooth — function g(.). This type of problems have recently attracted a great deal of interest, mainly due to their omnipresence in top-notch applications. Under this locally Lipschitz continuous gradient setting, we analyze the convergence behavior of proximal Newton schemes with the added twist of a probable presence of inexact evaluations. We prove attractive convergence rate guarantees and enhance state-of-the-art optimization schemes to accommodate such developments. Experimental results on sparse covariance estimation show the merits of our algorithm, both in terms of recovery efficiency and complexity.