Modern data analysis requires a number of tools to undercover hidden structure. For initial
exploration of data, animated scatter diagrams and nonparametric density estimation in …

In order to deal with the curse of dimensionality in reinforcement learning (RL), it is common
practice to make parametric assumptions where values or policies are functions of some low …

We advocate an optimization-centric view of Bayesian inference. Our inspiration is the
representation of Bayes' rule as infinite-dimensional optimization (Csisz´ r, 1975; Donsker …

We study high-dimensional distribution learning in an agnostic setting where an adversary is
allowed to arbitrarily corrupt an ε-fraction of the samples. Such questions have a rich history …

Let μ _N μ N be the empirical measure associated to a N N-sample of a given probability
distribution μ μ on R^ d R d. We are interested in the rate of convergence of μ _N μ N to μ μ …

We study the sample complexity of model-based reinforcement learning (henceforth RL) in
general contextual decision processes that require strategic exploration to find a near …

This book concerns the problem of data smoothing. There are many smoothing techniques,
yet the kernel smoothing seems to be one of the most important and widely used ones. In …

We make progress in a long-standing problem of batch reinforcement learning (RL):
learning Q* from an exploratory and polynomial-sized dataset, using a realizable and …

Children learn effortlessly by example and exhibit a remarkable capacity of generalization.
The field of machine learning, on the other hand, stumbles along clumsily in search of …

We provide a new computationally efficient class of estimators for risk minimization. We
show that these estimators are robust for general statistical models, under varied robustness …