My research focuses on the relationship between computation and statistics, with focus on applications in artificial intelligence, molecular biology and signal processing. I am particularly interested in the field of probabilistic graphical models, a field that links graph theory and probability theory in the design and analysis of statistical models, and kernel machines, a class of nonparametric methods based on computationally-efficient representations of generalized notions of similarity. In both cases, tools from optimization theory are indispensable, and my work also involves research in optimization theory.