Control Problems with Equality Constraints

not available at time of publication. Dimitris Bertsimas Massachusetts Insitute of Technology [email protected] MS8 Less is More: Robustness and Sparsity in Multivariate Statistics We describe applications of robust optimization in the context of machine learning, with a focus on classification and regression, principal component analysis and fitting of graphical models. In these tasks, sparsity of the solution is often desired as it provides ways to interpret the result. In a classification problem for example, the sparsity pattern of the classifier vector allows to identify the features that are most discriminative. We will explore the connections between sparsity and robustness and illustrate these connections in the context of genomic data, as well as social data, such as voting records and online news. El G. Laurent EECS, U.C. Berkeley [email protected] MS8 Robust Optimization in Multi-Period Problems We discuss computational tractability of Affinely Adjustable Robust Counterpart of an uncertain Linear Programming problem, with emphasis on a generic application, specifically, on computationally efficient design of affine control rules for an uncertainty-affected linear dynamical system ensuring that the state-control trajectory on a given finite horizon satisfies in a robust fashion a system of linear inequalities. Arkadi Nemirovski School of Industrial and Systems Engineering Georgia Tech [email protected]