Colloquium - Ka Wai (Raymond) Wong - October 09, 2025

Speaker: Ka Wai (Raymond) Wong, TAMU
Date/Time: Thursday, October 09, 2025, 10:00 AM - 11:00 AM ET

Title: A Principled Path to Fitted Distributional Evaluation

Abstract: In reinforcement learning, distributional off-policy evaluation (OPE) aims to estimate the return distribution of a target policy using offline data collected under a potentially different behavior policy. In this talk, I will focus on an approach called fitted distributional evaluation (FDE), which extends the widely used fitted Q-evaluation -- developed for expectation-based reinforcement learning -- to the distributional OPE setting. Although a few related methods exist, there is currently no unified framework for designing FDE algorithms. To address this, I will present a set of guiding principles for constructing theoretically sound FDE methods. Building on these principles, we can develop several new FDE algorithms with convergence guarantees. Moreover, this framework provides a theoretical foundation for existing methods, even in complex, non-tabular settings.

Bio: Raymond K. W. Wong is a Professor and PhD Program Director in the Department of Statistics at Texas A&M University. Prior to joining Texas A&M, he was a faculty member at Iowa State University. He earned his Ph.D. in Statistics from the University of California, Davis, in 2014. His research focuses on causal inference, functional data analysis, low-rank modeling, and reinforcement learning. Dr. Wong has served as an associate editor for several leading statistics journals, including the Canadian Journal of Statistics, Journal of Computational and Graphical Statistics, and Journal of the American Statistical Association, Review. He has also served as an area chair for major machine learning conferences such as ICML and NeurIPS.

Website: https://raymondkww.github.io/