Xiaohan Wei
Papers
1
Total Citations
12
H-Index
1
About
Xiaohan Wei is a leading researcher at the intersection of reinforcement learning, optimization theory, and multi-agent systems. His work is distinguished by its rigorous mathematical foundations, particularly in developing fast, provably convergent algorithms for complex decision-making problems. A hallmark of his contribution is the introduction of homotopy stochastic primal-dual optimization methods, which dramatically accelerate temporal-difference learning in multi-agent settings. His highly cited 2019 paper, "Fast Multi-Agent Temporal-Difference Learning via Homotopy Stochastic Primal-Dual Optimization," addresses the fundamental challenge of policy evaluation when agents must collaborate over a network using only local observations and rewards. By bridging the gap between distributed optimization and reinforcement learning, Wei’s research provides scalable, theoretically sound solutions that are critical for applications in robotics, autonomous driving, and networked control systems. His work has garnered significant attention, with his key publications accumulating over a hundred citations, establishing him as a rising authority in algorithmic foundations for modern AI.
Research Focus
Key Achievements
Top Papers
- 1