We consider the task of reinforcement learning with linear value function approximation. Temporal difference algorithms, and in particular the Least-Squares Temporal Difference (L...
We propose a framework for quantitative security analysis of machine learning methods. Key issus of this framework are a formal specification of the deployed learning model and a...
When correct priors are known, Bayesian algorithms give optimal decisions, and accurate confidence values for predictions can be obtained. If the prior is incorrect however, these...
Thomas Melluish, Craig Saunders, Ilia Nouretdinov,...
Learning the common structure shared by a set of supervised tasks is an important practical and theoretical problem. Knowledge of this structure may lead to better generalization ...
Andreas Argyriou, Charles A. Micchelli, Massimilia...
We propose a family of learning algorithms based on a new form of regularization that allows us to exploit the geometry of the marginal distribution. We focus on a semi-supervised...