In stochastic multi--armed bandits the objective is to solve the exploration--exploitation dilemma and ultimately maximize the expected reward. Nonetheless, in many practical problems, maximizing the expected reward is not the most desirable objective. In this paper, we introduce a novel setting based on the principle of risk--aversion where the objective is to compete against the arm with the best risk--return trade--off. This setting proves to be intrinsically more difficult than the standard multi-arm bandit setting due in part to an exploration risk which introduces a regret associated to the variability of an algorithm. Using variance as a measure of risk, we introduce two new algorithms, we investigate their theoretical guarantees, and we report preliminary empirical results.

Warped Gaussian processes (WGP) [1] model output observations in regression tasks as a parametric nonlinear transformation of a Gaussian process (GP). The use of this nonlinear transformation, which is included as part of the probabilistic model, was shown to enhance performance by providing a better prior model on several data sets. In order to learn its parameters, maximum likelihood was used. In this work we show that it is possible to use a non-parametric nonlinear transformation in WGP and variationally integrate it out. The resulting Bayesian WGP is then able to work in scenarios in which the maximum likelihood WGP failed: Low data regime, data with censored values, classification, etc. We demonstrate the superior performance of Bayesian warped GPs on several real data sets.

The partition function plays a key role in probabilistic modeling including conditional random fields, graphical models, and maximum likelihood estimation. To optimize partition functions, this article introduces a quadratic variational upper bound. This inequality facilitates majorization methods: optimization of complicated functions through the iterative solution of simpler sub-problems. Such bounds remain efficient to compute even when the partition function involves a graphical model (with small tree-width) or in latent likelihood settings. For large-scale problems, low-rank versions of the bound are provided and outperform LBFGS as well as first-order methods. Several learning applications are shown and reduce to fast and convergent update rules. Experimental results show advantages over state-of-the-art optimization methods.

We present a new learning strategy based on an efficient blocked Gibbs sampler for sparse overcomplete linear models. Particular emphasis is placed on statistical image modeling, where overcomplete models have played an important role in discovering sparse representations. Our Gibbs sampler is faster than general purpose sampling schemes while also requiring no tuning as it is free of parameters. Using the Gibbs sampler and a persistent variant of expectation maximization, we are able to extract highly sparse distributions over latent sources from data. When applied to natural images, our algorithm learns source distributions which resemble spike-and-slab distributions. We evaluate the likelihood and quantitatively compare the performance of the overcomplete linear model to its complete counterpart as well as a product of experts model, which represents another overcomplete generalization of the complete linear model. In contrast to previous claims, we find that overcomplete representations lead to significant improvements, but that the overcomplete linear model still underperforms other models.

In this work, we consider the problem of modeling the dynamic structure of human activities in the attributes space. A video sequence is first represented in a semantic feature space, where each feature encodes the probability of occurrence of an activity attribute at a given time. A generative model, denoted the binary dynamic system (BDS), is proposed to learn both the distribution and dynamics of different activities in this space. The BDS is a non-linear dynamic system, which extends both the binary principal component analysis (PCA) and classical linear dynamic systems (LDS), by combining binary observation variables with a hidden Gauss-Markov state process. In this way, it integrates the representation power of semantic modeling with the ability of dynamic systems to capture the temporal structure of time-varying processes. An algorithm for learning BDS parameters, inspired by a popular LDS learning method from dynamic textures, is proposed. A similarity measure between BDSs, which generalizes the Binet-Cauchy kernel for LDS, is then introduced and used to design activity classifiers. The proposed method is shown to outperform similar classifiers derived from the kernel dynamic system (KDS) and state-of-the-art approaches for dynamics-based or attribute-based action recognition.

We address the problem of estimating the difference between two probability densities. A naive approach is a two-step procedure of first estimating two densities separately and then computing their difference. However, such a two-step procedure does not necessarily work well because the first step is performed without regard to the second step and thus a small estimation error incurred in the first stage can cause a big error in the second stage. In this paper, we propose a single-shot procedure for directly estimating the density difference without separately estimating two densities. We derive a non-parametric finite-sample error bound for the proposed single-shot density-difference estimator and show that it achieves the optimal convergence rate. We then show how the proposed density-difference estimator can be utilized in L2-distance approximation. Finally, we experimentally demonstrate the usefulness of the proposed method in robust distribution comparison such as class-prior estimation and change-point detection.

Value Pursuit Iteration (VPI) is an approximate value iteration algorithm that finds a close to optimal policy for reinforcement learning and planning problems with large state spaces. VPI has two main features: First, it is a nonparametric algorithm that finds a good sparse approximation of the optimal value function given a dictionary of features. The algorithm is almost insensitive to the number of irrelevant features. Second, after each iteration of VPI, the algorithm adds a set of functions based on the currently learned value function to the dictionary. This increases the representation power of the dictionary in a way that is directly relevant to the goal of having a good approximation of the optimal value function. We theoretically study VPI and provide a finite-sample error upper bound for it.

Factor analysis models effectively summarise the covariance structure of high dimensional data,but the solutions are typically hard to interpret. This motivates attempting to find a disjoint partition, i.e. a simple clustering, of observed variables into highly correlated subsets. We introduce a Bayesian nonparametric approach to this problem, and demonstrate advantages over heuristic methods proposed to date. Our Dirichlet process variable clustering (DPVC) model can discover blockdiagonal covariancestructures in data. We evaluate our method on both synthetic and gene expression analysis problems.

Natural sounds exhibit complex statistical regularities at multiple scales. Acoustic eventsunderlying speech, for example, are characterized by precise temporal and frequency relationships, but they can also vary substantially according to the pitch, duration, and other high-level properties of speech production. Learning this structure from data while capturing the inherent variability is an important first step in building auditory processing systems, as well as understanding the mechanisms of auditory perception. Here we develop Hierarchical Spike Coding, a two-layer probabilistic generative model for complex acoustic structure. The first layer consists of a sparse spiking representation that encodes the sound using kernelspositioned precisely in time and frequency. Patterns in the positions of first layer spikes are learned from the data: on a coarse scale, statistical regularities areencoded by a second-layer spiking representation, while fine-scale structure is captured by recurrent interactions within the first layer. When fit to speech data, the second layer acoustic features include harmonic stacks, sweeps, frequency modulations, and precise temporal onsets, which can be composed to represent complex acoustic events. Unlike spectrogram-based methods, the model gives a probability distribution over sound pressure waveforms. This allows us to use the second-layer representation to synthesize sounds directly, and to perform model-based denoising, on which we demonstrate a significant improvement over standard methods.

In this paper, we provide a new framework to study the generalization bound of the learning process for domain adaptation. Without loss of generality, we consider two kinds of representative domain adaptation settings: one is domain adaptation with multiple sources and the other is domain adaptation combining source and target data. In particular, we introduce two quantities that capture the inherent characteristics of domains. For either kind of domain adaptation, based on the two quantities, we then develop the specific Hoeffding-type deviation inequality and symmetrization inequality to achieve the corresponding generalization bound based on the uniform entropy number. By using the resultant generalization bound, we analyze the asymptotic convergence and the rate of convergence of the learning process for such kind of domain adaptation. Meanwhile, we discuss the factors that affect the asymptotic behavior of the learning process. The numerical experiments support our results.