of Minnesota, and BS in math from Peking University. I was a postdoc at Stanford, obtained PhD from Univ. Before joining UIUC, I was a visiting scientist at FAIR (Facebook AI Research). Mathematical Programming, 171(1):115–166, 2018. Welcome to my homepage I am an assitant professor at UIUC, studying optimization and machine learning (especially deep learning). Data-driven distributionally robust optimization using the Wasserstein metric: performance guarantees and tractable reformulations. Known as a global leader in amino acids, ADM also offers, high-quality feed products, supplements, premixes, custom ingredient blends and specialty feed ingredients to aid in optimizing animal health and nutrition goals. Peyman Mohajerin Esfahani and Daniel Kuhn. ADM Animal Nutrition TM is a leading manufacturing, nutrition and marketing business offering a wide range of leading-edge products for the animal nutrition market. In recent years, there has been significant interest in ambiguity sets including all probability distributions within a given Wasserstein distance from a benchmark distribution (e.g., the empirical distribution defined by observed samples), mainly because of their computational tractability and of their attractive statistical guarantees see~\cite. Under this model, the decision-maker prepares for the worst-case distribution in the ambiguity set. This motivates us to formulate decision problems as a game: the decision-maker first chooses a decision then, some fictitious adversarial player (e.g., ``nature'') selects a probability distribution from a prescribed ambiguity set with the goal to inflict maximum harm. Many real-world decision problems arising in engineering and management have uncertain parameters, whose probability distributions is unknown. Topics in Distributionally Robust Optimization