My name is Xiaoyu Chen. I am a fifth-year PhD student in Peking University. I am very fortunate to be advised by Prof. Liwei Wang. Before that, I finished my undergraduate study with honors at Peking University. My research interests lie in reinforcement learning and online learning. For reinforcement learning, I am interested in designing efficient exploration algorithms with theoretical guarantees. For online learning, I mainly focus on bandit learning.

News

• One paper accepted by ICML 2023. April. 2023
• One paper accepted by ICML 2022. May. 2022
• Two papers accepted by ICLR 2022 (spotlight). Feb. 2022

Selected Publications

• (ICLR 2022 spotlight) Understanding Domain Randomization for Sim-to-real Transfer
  Xiaoyu Chen*, Jiachen Hu*, Chi Jin, Lihong Li, Liwei Wang
  Highlight: We model the simulator in sim-to-real transfer as a set of MDPs with tunable parameters. We prove that zero-shot sim-to-real transfer can succeed under mild assumptions if we use domain randomization methods with a memory augmented policy model.

• (ICLR 2022 spotlight) Near-Optimal Reward-Free Exploration for Linear Mixture MDPs with Plug-in Solver
  Xiaoyu Chen, Jiachen Hu, Lin F. Yang, Liwei Wang
  Highlight: We formulate the reward-free setting with a plug-in solver, in which the agent learns a model in the exploration phase and demands that any planning algorithm on the learned model can give a near-optimal policy. We provide efficient learning algorithms with near-optimal sample complexity bounds.