Haolin Liu

Department of Computer Science, University of Virginia

Hi! I am Haolin Liu, a third-year PhD student at the University of Virginia, where I am fortunate to be advised by Prof. Chen-Yu Wei. Prior to this, I received my bachelor’s degree in Computer Science from ShanghaiTech University, where I studied chemistry for 1.5 years before transitioning to computer science for 2.5 years. I finished my undergraduate thesis with Prof. Dengji Zhao.

My research is centered on Reinforcement Learning (RL), spanning both theoretical and practical domains.

  • On the theoretical side, I am dedicated to advancing our understanding of how RL algorithms can effectively learn and adapt in dynamic environments.
  • Practically, I specialize in applying RL to the post-training of large language models (LLMs), with a particular emphasis on alignment and reasoning.

I also have some experience in game theory and mechanism design.

selected publications

  1. NeurIPS
    Beating Adversarial Low-Rank MDPs with Unknown Transition and Bandit Feedback
    (α-β) Haolin Liu, Zakaria Mhammedi, Chen-Yu Wei, and Julian Zimmert
    NeurIPS, 2024
  2. NeurIPS
    Corruption-Robust Linear Bandits: Minimax Optimality and Gap-Dependent Misspecification
    (α-β) Haolin Liu, Artin Tajdini, Andrew Wagenmaker, and Chen-Yu Wei
    NeurIPS, 2024
  3. ICLR
    Towards Optimal Regret in Adversarial Linear MDPs with Bandit Feedback (Spotlight)
    (α-β) Haolin Liu, Chen-Yu Wei, and Julian Zimmert
    ICLR, 2024
  4. NeurIPS
    Bypassing the simulator: Near-optimal adversarial linear contextual bandits
    (α-β) Haolin Liu, Chen-Yu Wei, and Julian Zimmert
    NeurIPS, 2023