Prof. Enlu Zhou
Georgia Institute of Technology
Talk:
Bayesian Approaches to Data-driven Stochastic Optimization
Abstract:
A large class of stochastic optimization problems involves optimizing an expectation taken with respect to an underlying distribution that is unknown in practice. One popular approach to addressing distributional uncertainty, known as the distributionally robust optimization (DRO), is to hedge against the worst case among an ambiguity set of candidate distributions. However, given that the worst case rarely happens, inappropriate construction of the ambiguity set can sometimes result in overly conservative solutions. We introduce new formulations that utilize Bayesian posterior distribution to characterize uncertainty in the estimated distribution and adopt risk measures to allow a more flexible attitude towards risk. Our approaches apply to a wide class of data-driven stochastic optimization problems, including static and dynamic settings, and exogenous (decision-independent) and endogenous (decision-dependent) uncertainties.
Biography:
Enlu Zhou is a Professor in the H. Milton Stewart School of Industrial & Systems Engineering at Georgia Institute of Technology. She received the B.S. degree with highest honors in electrical engineering from Zhejiang University, China, in 2004, and the Ph.D. degree in electrical engineering from the University of Maryland, College Park, in 2009. Prior to joining Georgia Tech, she was an assistant professor in the Industrial & Enterprise Systems Engineering Department at the University of Illinois Urbana-Champaign from 2009 to 2013. She is a recipient of the AFOSR Young Investigator award, NSF CAREER award, the INFORMS Outstanding Simulation Publication award, and the Best Theoretical Paper award at the Winter Simulation Conference (twice). She has served as an associate editor for Journal of Simulation, IEEE Transactions on Automatic Control, and Operations Research. Currently, she is the Vice President and President-Elect of the INFORMS Simulation Society.
