About Me
I am an Assistant Professor in the Daniel J. Epstein Department of Industrial and Systems Engineering (ISE) at the University of Southern California. I received my Ph.D. from the Department of Management Science and Engineering (MS&E) at Stanford University, where I was fortunate to be co-advised by Prof. Peter Glynn and Prof. Jose Blanchet. Prior to my doctoral studies at Stanford, I completed my B.S. degree at Cornell Engineering, majoring in Operations Research and Information Engineering (ORIE).
Research Interests
I am interested in a wide range of research areas at the intersection of applied probability, reinforcement learning (RL), machine learning, and simulation. My work focuses on the foundation, design, and analysis of algorithms for learning and controlling dynamic engineering systems, with applications in management science and operations research. In particular, I address the reliability and scalability challenges that arise in contemporary problems. Key areas of my research include:
- Developing efficient RL and approximate dynamic programming algorithms for policy learning and optimization with provable performance guarantees.
- Establishing theoretical foundations for robust Markov decision processes and distributionally robust stochastic control.
- Building statistically tractable data-driven models and efficient algorithms for reliable RL.
- Advancing deep learning techniques for policy learning by leveraging simulation methodologies and applied probabilistic tools.
Research Highlights
I am excited to share my recent work on Q-Measure-Learning, a novel and efficiently implementable algorithm for continuous-state RL.
- Q-Measure-Learning for Continuous State RL: Efficient Implementation and Convergence
- Shengbo Wang
- Some slides for this work
We are pleased to present our latest findings on the dynamic programming theory for average-reward robust Markov decision processes in the following new manuscript.