Qiwen Cui

Qiwen Cui Ph.D. student Paul G. Allen School of Computer Science & Engineering University of Washington Email: qwcui at cs (dot) washington (dot) edu Google Scholar

About me

I am a final year Ph.D. student in the Paul G. Allen School of Computer Science & Engineering at the University of Washington. I am very fortunate to be advised by Professor Simon Shaolei Du. Prior to starting my Ph.D. study, I did my undergrad in the School of Mathematical Sciences at Peking University advised by Professor Zaiwen Wen. I had a great summer working with Professor Lin F. Yang in 2020, who led me into the world of reinforcement learning theory.

My research interests are in reinforcement learning. I have been working on designing provably efficient multi-agent reinforcement learning algorithms for the offline case (exploiting the dataset) and the online case (exploring the structured environment). I have also done research in optimization and game theory. I am also interested in the role of RL(HF) in the performance of LLMs.

Selected Publications

Offline Multi-agent Reinforcement Learning

• Multi-Agent Reinforcement Learning from Human Feedback: Data Coverage and Algorithmic Techniques
  Natalia Zhang*, Xinqi Wang*, Qiwen Cui*, Runlong Zhou, Sham M Kakade, Simon S Du
  ArXiv: 2409.00717

• Provably Efficient Offline Multi-agent Reinforcement Learning via Strategy-wise Bonus
  Qiwen Cui, Simon S. Du
  Conference on Neural Information Processing Systems (NeurIPS) 2022
  

• When is Offline Two-Player Zero-Sum Markov Game Solvable?
  Qiwen Cui, Simon S. Du
  Conference on Neural Information Processing Systems (NeurIPS) 2022
  

Online Multi-agent Reinforcement Learning

• Refined Sample Complexity for Markov Games with Independent Linear Function Approximation
  Yan Dai, Qiwen Cui, Simon S. Du
  The 37th Annual Conference on Learning Theory (COLT) 2024
  

• A Black-box Approach for Non-stationary Multi-agent Reinforcement Learning
  Haozhe Jiang, Qiwen Cui, Zhihan Xiong, Maryam Fazel, Simon S. Du
  International Conference on Learning Representations (ICLR) 2024
  

• Breaking the Curse of Multiagents in a Large State Space: RL in Markov Games with Independent Linear Function Approximation
  Qiwen Cui, Kaiqing Zhang, Simon S. Du
  The 36th Annual Conference on Learning Theory (COLT) 2023
  

Learning in Congestion Games

• Learning Optimal Tax Design in Nonatomic Congestion Games
  Qiwen Cui, Maryam Fazel, Simon S. Du
  Conference on Neural Information Processing Systems (NeurIPS) 2024
  

• Offline Congestion Games: How Feedback Type Affects Data Coverage Requirement
  Haozhe Jiang*, Qiwen Cui*, Zhihan Xiong, Maryam Fazel, Simon S. Du
  International Conference on Learning Representations (ICLR) 2023
  

• Learning in Congestion Games with Bandit Feedback
  Qiwen Cui*, Zhihan Xiong*, Maryam Fazel, Simon S. Du
  Conference on Neural Information Processing Systems (NeurIPS) 2022
  

Reinforcement Learning for LLM

• BabelBench: An Omni Benchmark for Code-Driven Analysis of Multimodal and Multistructured Data
  Xuwu Wang, Qiwen Cui, Yunzhe Tao, Yiran Wang, Ziwei Chai, Xiaotian Han, Boyi Liu, Jianbo Yuan, Jing Su, Guoyin Wang, Tingkai Liu, Liyu Chen, Tianyi Liu, Tao Sun, Yufeng Zhang, Sirui Zheng, Quanzeng You, Yang Yang, Hongxia Yang
  ArXiv: 2410.00773
  

• (N,K)-Puzzle: A Cost-Efficient Testbed for Benchmarking Reinforcement Learning Algorithms in Generative Language Model
  Yufeng Zhang, Liyu Chen, Boyi Liu, Yingxiang Yang, Qiwen Cui, Yunzhe Tao, Hongxia Yang
  ArXiv: 2403.07191