Zaiwei Chen
I am an Assistant Professor in the School of Industrial Engineering at Purdue University.
Previously, I was a CMI Postdoctoral Fellow at Caltech in the Computing and Mathematical Sciences Department, working with Dr. Adam Wierman and Dr. Eric Mazumdar. I hold a Ph.D. in Machine Learning, an M.S. in Mathematics, and an M.S. in Operations Research from Georgia Tech, where I was advised by Dr. Siva Theja Maguluri and Dr. John-Paul Clarke. I earned my B.S. in Electrical Engineering from Chu Kochen Honors College, Zhejiang University.
I was appointed as a Solberg Academic Excellence Scholar in the School of Industrial Engineering at Purdue University. During my postdoctoral studies, I received the Simoudis Discovery Prize and was named a PIMCO Postdoctoral Fellow in Data Science in 2022. My Ph.D. thesis won the Sigma Xi Best Ph.D. Thesis Award and was a runner-up for the 2022 SIGMETRICS Doctoral Dissertation Award. Additionally, during my Ph.D., I was awarded the ARC-TRIAD Student Fellowship in 2021 and the IDEaS-TRIAD Research Scholarship in 2020.
Motivated by real-world challenges and limitations, my research interest is broadly in the development of theoretical foundations of sequential decision-making under uncertainty. More specifically, I am focusing on designing data-efficient (multi-agent) reinforcement learning algorithms that are substantiated with theoretical guarantees. Here is my CV.
I am actively looking for motivated Ph.D. students with strong mathematical backgrounds. Interested candidates should apply to the Ph.D. program in the School of Industrial Engineering at Purdue University.
Recent News
I will jointly chair an APS session titled "Strategic and Distributionally Robust Sequential Decision Making" at the 2024 INFORMS Annual Meeting.
I will give a talk titled "Approximate Global Convergence of Independent Learning in Multi-Agent Systems" in an Invited Session at the 2024 INFORMS Annual Meeting.
Our paper titled "Two-Timescale Q-Learning with Function Approximation in Zero-Sum Markov Games" was accepted by The 25th ACM Conference on Economics and Computation.
I gave a talk titled "Two-Timescale Q-Learning with Function Approximation in Zero-Sum Markov Games" at the 2024 INFORMS Optimization Society Conference.