Publications

(α-β order) denotes alphabetical authorship ordering, and (*,+) denote equal contribution

Recent Works

• Partially Observable RL: Benign Structures and Simple Generic Algorithms (Survey Article)
  Qinghua Liu, Chi Jin
  Invited Submission to Statistical Science (STS), 2024

• Learning to Achieve Goals with Belief State Transformers
  Edward S. Hu, Kwangjun Ahn, Qinghua Liu, Haoran Xu, Manan Tomar, Ada Langford,
  Dinesh Jayaraman, Alex Lamb, John Langford
  arXiv preprint

Partially Observable Reinforcement Learning

• Optimistic MLE – A Generic Model-based Algorithm for Partially Observable Sequential Decision Making
  Qinghua Liu, Praneeth Netrapalli, Csaba Szepesvári, Chi Jin
  Symposium on Theory of Computing (STOC), 2023

• Sample-Efficient Reinforcement Learning of Partially Observable Markov Games
  Qinghua Liu, Csaba Szepesvári, Chi Jin
  Neural Information Processing Systems (NeurIPS), 2022
  European Workshop on Reinforcement Learning, 2022 (Oral)

• When Is Partially Observable Reinforcement Learning Not Scary?
  Qinghua Liu, Alan Chung, Csaba Szepesvári, Chi Jin
  Conference on Learning Theory (COLT), 2022

• Sample-Efficient Reinforcement Learning of Undercomplete POMDPs
  (α-β order) Chi Jin, Sham M. Kakade, Akshay Krishnamurthy, Qinghua Liu
  Neural Information Processing Systems (NeurIPS), 2020 (Spotlight) [Slides] [RL Theory Seminar]

Multi-Agent Reinforcement Learning

• Breaking the Curse of Multiagency: Provably Efficient Decentralized MARL with Function Approximation
  Yuanhao Wang*, Qinghua Liu*, Yu Bai+, Chi Jin+
  Conference on Learning Theory (COLT), 2023

• Policy Optimization for Markov Games: Unified Framework and Faster Convergence
  Runyu Zhang*, Qinghua Liu*, Huan Wang, Caiming Xiong, Na Li, Yu Bai
  Neural Information Processing Systems (NeurIPS), 2022

• Learning Markov Games with Adversarial Opponents: Efficient Algorithms and Fundamental Limits
  Qinghua Liu*, Yuanhao Wang*, Chi Jin
  International Conference on Machine Learning (ICML), 2022 (Long oral)

• V-Learning – A Simple, Efficient, Decentralized Algorithm for Multiagent RL
  (α-β order) Chi Jin, Qinghua Liu, Yuanhao Wang, Tiancheng Yu
  Best Paper in ICLR Workshop on Gamification and Multiagent Solutions, 2022
  Mathematics of Operations Research, 2023
  

• A Deep Reinforcement Learning Approach for Finding Non-Exploitable Strategies in Two-Player Atari Games
  Zihan Ding, Dijia Su, Qinghua Liu, Chi Jin
  arXiv preprint

• The Power of Exploiter: Provable Multi-Agent RL in Large State Spaces
  (α-β order) Chi Jin, Qinghua Liu, Tiancheng Yu
  International Conference on Machine Learning (ICML), 2022
  ICML Workshop on Reinforcement Learning Theory, 2021
  

• A Sharp Analysis of Model-based Reinforcement Learning with Self-Play
  Qinghua Liu, Tiancheng Yu, Yu Bai, Chi Jin
  International Conference on Machine Learning (ICML), 2021

Reinforcement Learning with Large State Spaces

• Is RLHF More Difficult than Standard RL?
  Yuanhao Wang, Qinghua Liu, Chi Jin
  Neural Information Processing Systems (NeurIPS), 2023

• Optimistic Natural Policy Gradient: a Simple Efficient Policy Optimization Framework for Online RL
  Qinghua Liu, Gellért Weisz, András György, Chi Jin, Csaba Szepesvári
  Neural Information Processing Systems (NeurIPS), 2023 (Spotlight)

• Bellman Eluder Dimension: New Rich Classes of RL Problems, and Sample-Efficient Algorithms
  (α-β order) Chi Jin, Qinghua Liu, Sobhan Miryoosefi
  Neural Information Processing Systems (NeurIPS), 2021 (Spotlight) [Slides] [RL Theory Seminar]

• Provable Rich Observation Reinforcement Learning with Combinatorial Latent States
  Dipendra Misra, Qinghua Liu, Chi Jin, John Langford
  International Conference on Learning Representations (ICLR), 2021

Other Works

• Context-lumpable Stochastic Bandits
  Chung-Wei Lee, Qinghua Liu, Yasin Abbasi-Yadkori, Chi Jin, Tor Lattimore, Csaba Szepesvári
  Neural Information Processing Systems (NeurIPS), 2023

• Tackling the Objective Inconsistency Problem in Heterogeneous Federated Optimization
  Jianyu Wang, Qinghua Liu, Hao Liang, Gauri Joshi, H. Vincent Poor
  Neural Information Processing Systems (NeurIPS), 2020
  Longer version in IEEE Transactions on Signal Processing

• On Limitation of Transformer for Learning HMMs
  Jiachen Hu, Qinghua Liu, Chi Jin
  arXiv preprint

• A Tight Lower Bound for Uniformly Stable Algorithms
  (α-β order) Qinghua Liu, Zhou Lu
  arXiv preprint