Our research advances how machines can learn, predict or control, and do so at scale in an efficient, principled, and interpretable manner. Our research in machine learning extends from foundational theory to modern applications, focusing especially on statistical inference and estimation tasks that lie at the heart of complex learning problems. We design new methods, theory and algorithms so as to automate the use and generation of semi-structured data such as natural language text, images, molecules, or strategies. We apply and develop our algorithms to solve multi-faceted recommender, retrieval, or inferential tasks (e.g., biomedical), design and optimize molecules or reactions for the purpose of drug design, and to model strategic, game theoretic interactions.