Matching pursuit (MP) methods are a promising class of feature construction algorithms for value function approximation. Yet existing MP methods require creating a pool of potential features, mandating expert knowledge or enumeration of a large feature pool, both of which hinder scalability. This paper introduces batch incremental feature dependency discovery (Batch-iFDD) as an MP method that inherits a provable convergence property. Additionally, Batch-iFDD does not require a large pool of features, leading to lower computational complexity. Empirical policy evaluation results across three domains with up to one million states highlight the scalability of Batch-iFDD over the previous state of the art MP algorithm.

We present a comprehensive study of the use of generative modeling approaches for Multiple-Instance Learning (MIL) problems. In MIL a learner receives training instances grouped together into bags with labels for the bags only (which might not be correct for the comprised instances). Our work was motivated by the task of facilitating the diagnosis of neuromuscular disorders using sets of motor unit potential trains (MUPTs) detected within a muscle which can be cast as a MIL problem. Our approach leads to a state-of-the-art solution to the problem of muscle classification. By introducing and analyzing generative models for MIL in a general framework and examining a variety of model structures and components, our work also serves as a methodological guide to modelling MIL tasks. We evaluate our proposed methods both on MUPT datasets and on the MUSK1 dataset, one of the most widely used benchmarks for MIL.

This paper addresses the problem of rank aggregation, which aims to find a consensus ranking among multiple ranking inputs. Traditional rank aggregation methods are deterministic, and can be categorized into explicit and implicit methods depending on whether rank information is explicitly or implicitly utilized. Surprisingly, experimental results on real data sets show that explicit rank aggregation methods would not work as well as implicit methods, although rank information is critical for the task. Our analysis indicates that the major reason might be the unreliable rank information from incomplete ranking inputs. To solve this problem, we propose to incorporate uncertainty into rank aggregation and tackle the problem in both unsupervised and supervised scenario. We call this novel framework {stochastic rank aggregation} (St.Agg for short). Specifically, we introduce a prior distribution on ranks, and transform the ranking functions or objectives in traditional explicit methods to their expectations over this distribution. Our experiments on benchmark data sets show that the proposed St.Agg outperforms the baselines in both unsupervised and supervised scenarios.

Conventional learning with expert advice methods assumes a learner is always receiving the outcome (e.g., class labels) of every incoming training instance at the end of each trial. In real applications, acquiring the outcome from oracle can be costly or time consuming. In this paper, we address a new problem of active learning with expert advice, where the outcome of an instance is disclosed only when it is requested by the online learner. Our goal is to learn an accurate prediction model by asking the oracle the number of questions as small as possible. To address this challenge, we propose a framework of active forecasters for online active learning with expert advice, which attempts to extend two regular forecasters, i.e., Exponentially Weighted Average Forecaster and Greedy Forecaster, to tackle the task of active learning with expert advice. We prove that the proposed algorithms satisfy the Hannan consistency under some proper assumptions, and validate the efficacy of our technique by an extensive set of experiments.

We present a novel approach for constrained Bayesian inference. Unlike current methods, our approach does not require convexity of the constraint set. We reduce the constrained variational inference to a parametric optimization over the feasible set of densities and propose a general recipe for such problems. We apply the proposed constrained Bayesian inference approach to multitask learning subject to rank constraints on the weight matrix. Further, constrained parameter estimation is applied to recover the sparse conditional independence structure encoded by prior precision matrices. Our approach is motivated by reverse inference for high dimensional functional neuroimaging, a domain where the high dimensionality and small number of examples requires the use of constraints to ensure meaningful and effective models. For this application, we propose a model that jointly learns a weight matrix and the prior inverse covariance structure between different tasks. We present experimental validation showing that the proposed approach outperforms strong baseline models in terms of predictive performance and structure recovery.

Graph-based methods provide a powerful tool set for many non-parametric frameworks in Machine Learning. In general, the memory and computational complexity of these methods is quadratic in the number of examples in the data which makes them quickly infeasible for moderate to large scale datasets. A significant effort to find more efficient solutions to the problem has been made in the literature. One of the state-of-the-art methods that has been recently introduced is the Variational Dual-Tree (VDT) framework. Despite some of its unique features, VDT is currently restricted only to Euclidean spaces where the Euclidean distance quantifies the similarity. In this paper, we extend the VDT framework beyond the Euclidean distance to more general Bregman divergences that include the Euclidean distance as a special case. By exploiting the properties of the general Bregman divergence, we show how the new framework can maintain all the pivotal features of the VDT framework and yet significantly improve its performance in non-Euclidean domains. We apply the proposed framework to different text categorization problems and demonstrate its benefits over the original VDT.

A number of discrete and continuous optimization problems in machine learning are related to convex minimization problems under submodular constraints. In this paper, we deal with a submodular function with a directed graph structure, and we show that a wide range of convex optimization problems under submodular constraints can be solved much more efficiently than general submodular optimization methods by a reduction to a maximum flow problem. Furthermore, we give some applications, including sparse optimization methods, in which the proposed methods are effective. Additionally, we evaluate the performance of the proposed method through computational experiments.

We consider the problem of maximum a posteriori (MAP) inference in discrete graphical models. We present a parallel MAP inference algorithm called Bethe-ADMM based on two ideas: tree-decomposition of the graph and the alternating direction method of multipliers (ADMM). However, unlike the standard ADMM, we use an inexact ADMM augmented with a Bethe-divergence based proximal function, which makes each subproblem in ADMM easy to solve in parallel using the sum-product algorithm. We rigorously prove global convergence of Bethe-ADMM. The proposed algorithm is extensively evaluated on both synthetic and real datasets to illustrate its effectiveness. Further, the parallel Bethe-ADMM is shown to scale almost linearly with increasing number of cores.

We consider the problem of providing nonparametric confidence guarantees for undirected graphs under weak assumptions. In particular, we do not assume sparsity, incoherence or Normality. We allow the dimension $D$ to increase with the sample size $n$. First, we prove lower bounds that show that if we want accurate inferences with low assumptions then there are limitations on the dimension as a function of sample size. When the dimension increases slowly with sample size, we show that methods based on Normal approximations and on the bootstrap lead to valid inferences and we provide Berry-Esseen bounds on the accuracy of the Normal approximation. When the dimension is large relative to sample size, accurate inferences for graphs under low assumptions are not possible. Instead we propose to estimate something less demanding than the entire partial correlation graph. In particular, we consider: cluster graphs, restricted partial correlation graphs and correlation graphs.

In this paper, we present the Bennett-type generalization bounds of the learning process for i.i.d. samples, and then show that the generalization bounds have a faster rate of convergence than the traditional results. In particular, we first develop two types of Bennett-type deviation inequality for the i.i.d. learning process: one provides the generalization bounds based on the uniform entropy number; the other leads to the bounds based on the Rademacher complexity. We then adopt a new method to obtain the alternative expressions of the Bennett-type generalization bounds, which imply that the bounds have a faster rate o(N^{-1/2}) of convergence than the traditional results O(N^{-1/2}). Additionally, we find that the rate of the bounds will become faster in the large-deviation case, which refers to a situation where the empirical risk is far away from (at least not close to) the expected risk. Finally, we analyze the asymptotical convergence of the learning process and compare our analysis with the existing results.