报告题目:On Generalization and Regularization via Wasserstein Distributionally Robust Optimization

报告内容: Wasserstein distributionally robust optimization (DRO) has found success in operations research and machine learning applications as a powerful means to obtain solutions with favourable out-of-sample performances. Two compelling explanations for the success are the generalization bounds derived from Wasserstein DRO and the equivalency between Wasserstein DRO and the regularization scheme commonly applied in machine learning. We show that it is possible to obtain generalization bounds and the equivalency to regularization in a significantly broader setting where the Wasserstein ball can be of a general type and the decision criterion can be a general measure of risk, i.e., nonlinear in distributions. This allows for accommodating many important classification, regression, and risk minimization applications that have not been addressed to date using Wasserstein DRO. Our results are strong in that the generalization bounds do not suffer from the curse of dimensionality and the equivalency to regularization is exact. As a byproduct, our regularization results broaden considerably the class of Wasserstein DRO models that can be solved efficiently via regularization formulations.

报告人简介: Jonathan Yu-Meng Li is an Associate Professor in Risk Analytics at the Telfer School of Management, University of Ottawa. He earned his PhD in Operations Research and Financial Engineering from the University of Toronto. His research advances analytical methodologies and applications in optimization, machine learning, and financial econometrics. His methodological works on distributionally robust optimization, preference robust optimization, and inverse optimization have been featured in the journals Management Science and Operations Research. He spearheaded the business analytics graduate program and currently serves as the principal coordinator for the Center of Responsible Wealth Transition.