Efficient global optimization is the problem of minimizing an unknown function f, using as few evaluations f(x) as possible. It can be considered as a continuum-armed bandit problem, with noiseless data and simple regret. Expected improvement is perhaps the most popular method for solving this problem; the algorithm performs well in experiments, but little is known about its theoretical properties. Implementing expected improvement requires a choice of Gaussian process prior, which determines an associated space of functions, its reproducing-kernel Hilbert space (RKHS). When the prior is fixed, expected improvement is known to converge on the minimum of any function in the RKHS. We begin by providing convergence rates for this procedure. The rates are optimal for functions of low smoothness, and we modify the algorithm to attain optimal rates for smoother functions. For practitioners, however, these results are somewhat misleading. Priors are typically not held fixed, but depend on parameters estimated from the data. For standard estimators, we show this procedure may never discover the minimum of f. We then propose alternative estimators, chosen to minimize the constants in the rate of convergence, and show these estimators retain the convergence rates of a fixed prior.

In many professional sports leagues, teams from opposing leagues/conferences compete against one another, playing inter-league games. This is an example of a bipartite tournament. In this paper, we consider the problem of reducing the total travel distance of bipartite tournaments, by analyzing inter-league scheduling from the perspective of discrete optimization. This research has natural applications to sports scheduling, especially for leagues such as the National Basketball Association (NBA) where teams must travel long distances across North America to play all their games, thus consuming much time, money, and greenhouse gas emissions. We introduce the Bipartite Traveling Tournament Problem (BTTP), the inter-league variant of the well-studied Traveling Tournament Problem. We prove that the 2n-team BTTP is NP-complete, but for small values of n, a distance-optimal inter-league schedule can be generated from an algorithm based on minimum-weight 4-cycle-covers. We apply our theoretical results to the 12-team Nippon Professional Baseball (NPB) league in Japan, producing a provably-optimal schedule requiring 42950 kilometres of total team travel, a 16% reduction compared to the actual distance traveled by these teams during the 2010 NPB season. We also develop a nearly-optimal inter-league tournament for the 30-team NBA league, just 3.8% higher than the trivial theoretical lower bound.

Satisfied customers establishes loyalty, provides opportunities of selling additional products and services. Satisfied customers also reduce the probability of losing business to competitors. However, customer dissatisfaction results in direct revenue losses due to customer churn as well as damage to business reputation. Therefore, the improvement of customer experience is a vital priority for contact centres across all industries. Interactive Voice Response (IVR) systems are used by businesses to provide customers with a convenient, consistent and reliable contact channel to access information fast.

Abstract: We consider the problem of covariance matrix estimation in the presence of latent variables. Under suitable conditions, it is possible to learn the marginal covariance matrix of the observed variables via a tractable convex program, where the concentration matrix of the observed variables is decomposed into a sparse matrix (representing the graphical structure of the observed variables) and a low rank matrix (representing the marginalization effect of latent variables). We present an efficient first-order method based on split Bregman to solve the convex problem. The algorithm is guaranteed to converge under mild conditions. We show that our algorithm is significantly faster than the state-of-the-art algorithm on both artificial and real-world data. Applying the algorithm to a gene expression data involving thousands of genes, we show that most of the correlation between observed variables can be explained by only a few dozen latent factors.

The QXOR-SAT problem is the quantified version of the satisfiability problem XOR-SAT in which the connective exclusive-or is used instead of the usual or. We study the phase transition associated with random QXOR-SAT instances. We give a description of this phase transition in the case of one alternation of quantifiers, thus performing an advanced practical and theoretical study on the phase transition of a quantified problem.

Allocating scarce resources among agents to maximize global utility is, in general, computationally challenging. We focus on problems where resources enable agents to execute actions in stochastic environments, modeled as Markov decision processes (MDPs), such that the value of a resource bundle is defined as the expected value of the optimal MDP policy realizable given these resources. We present an algorithm that simultaneously solves the resource-allocation and the policy-optimization problems. This allows us to avoid explicitly representing utilities over exponentially many resource bundles, leading to drastic (often exponential) reductions in computational complexity. We then use this algorithm in the context of self-interested agents to design a combinatorial auction for allocating resources. We empirically demonstrate the effectiveness of our approach by showing that it can, in minutes, optimally solve problems for which a straightforward combinatorial resource-allocation technique would require the agents to enumerate up to 2^100 resource bundles and the auctioneer to solve an NP-complete problem with an input of that size.

We introduce a new wavelet transform suitable for analyzing functions on point clouds and graphs. Our construction is based on a generalization of the average interpolating refinement scheme of Donoho. The most important ingredient of the original scheme that needs to be altered is the choice of the interpolant. Here, we define the interpolant as the minimizer of a smoothness functional, namely a generalization of the Laplacian energy, subject to the averaging constraints. In the continuous setting, we derive a formula for the optimal solution in terms of the poly-harmonic Green's function. The form of this solution is used to motivate our construction in the setting of graphs and point clouds. We highlight the empirical convergence of our refinement scheme and the potential applications of the resulting wavelet transform through experiments on a number of data stets.

Choice models, which capture popular preferences over objects of interest, play a key role in making decisions whose eventual outcome is impacted by human choice behavior. In most scenarios, the choice model, which can effectively be viewed as a distribution over permutations, must be learned from observed data. The observed data, in turn, may frequently be viewed as (partial, noisy) information about marginals of this distribution over permutations. As such, the search for an appropriate choice model boils down to learning a distribution over permutations that is (near-)consistent with observed information about this distribution. In this work, we pursue a non-parametric approach which seeks to learn a choice model (i.e. a distribution over permutations) with {\em sparsest} possible support, and consistent with observed data. We assume that the data observed consists of noisy information pertaining to the marginals of the choice model we seek to learn. We establish that {\em any} choice model admits a `very' sparse approximation in the sense that there exists a choice model whose support is small relative to the dimension of the observed data and whose marginals approximately agree with the observed marginal information. We further show that under, what we dub, `signature' conditions, such a sparse approximation can be found in a computationally efficiently fashion relative to a brute force approach. An empirical study using the American Psychological Association election data-set suggests that our approach manages to unearth useful structural properties of the underlying choice model using the sparse approximation found. Our results further suggest that the signature condition is a potential alternative to the recently popularized Restricted Null Space condition for efficient recovery of sparse models.

We consider a class of learning problems regularized by a structured sparsity-inducing norm defined as the sum of l_2- or l_infinity-norms over groups of variables. Whereas much effort has been put in developing fast optimization techniques when the groups are disjoint or embedded in a hierarchy, we address here the case of general overlapping groups. To this end, we present two different strategies: On the one hand, we show that the proximal operator associated with a sum of l_infinity-norms can be computed exactly in polynomial time by solving a quadratic min-cost flow problem, allowing the use of accelerated proximal gradient methods. On the other hand, we use proximal splitting techniques, and address an equivalent formulation with non-overlapping groups, but in higher dimension and with additional constraints. We propose efficient and scalable algorithms exploiting these two strategies, which are significantly faster than alternative approaches. We illustrate these methods with several problems such as CUR matrix factorization, multi-task learning of tree-structured dictionaries, background subtraction in video sequences, image denoising with wavelets, and topographic dictionary learning of natural image patches.

Boolean optimization finds a wide range of application domains, that motivated a number of different organizations of Boolean optimizers since the mid 90s. Some of the most successful approaches are based on iterative calls to an NP oracle, using either linear search, binary search or the identification of unsatisfiable sub-formulas. The increasing use of Boolean optimizers in practical settings raises the question of confidence in computed results. For example, the issue of confidence is paramount in safety critical settings. One way of increasing the confidence of the results computed by Boolean optimizers is to develop techniques for validating the results. Recent work studied the validation of Boolean optimizers based on branch-and-bound search. This paper complements existing work, and develops methods for validating Boolean optimizers that are based on iterative calls to an NP oracle. This entails implementing solutions for validating both satisfiable and unsatisfiable answers from the NP oracle. The work described in this paper can be applied to a wide range of Boolean optimizers, that find application in Pseudo-Boolean Optimization and in Maximum Satisfiability. Preliminary experimental results indicate that the impact of the proposed method in overall performance is negligible.