Chuanhao Li is an Assistant Professor in the Department of Industrial Engineering at Tsinghua University. He received bachelor’s degrees in English and Mechanical Engineering, as well as a master’s degree in Mechatronics Engineering, from Harbin Institute of Technology, and a Ph.D. in Computer Science from the University of Virginia. His research focuses on learning and decision-making in multi-agent systems, aiming to improve system efficiency and performance across multiple dimensions through the design of learning and optimization algorithms.

2025.07 – Present Assistant Professor, Department of Industrial Engineering, Tsinghua University

2023.10 – 2025.06 Postdoctoral Associate, Department of Statistics and Data Science, Yale University

2018.08 – 2023.08 Ph.D. in Computer Science, University of Virginia, USA

2016.09 – 2018.07 M.S. in Mechatronics Engineering, Harbin Institute of Technology, China

2012.08 – 2016.07 B.A. in English and B.S. in Mechanical Engineering, Harbin Institute of Technology, China

I recruit Ph.D. students, master’s students, and undergraduate research assistants annually. Interested candidates are encouraged to send their CV to chuanhao-li{at}tsinghua.edu.cn.

What We Look For: We welcome self-motivated, curious, and communicative students to join our group. Candidates should have a solid foundation in mathematics and programming, along with theoretical or applied expertise in areas such as machine learning, reinforcement learning, or operations research.

I am interested in multi‑agent systems that reason, interact, and learn in dynamic and uncertain settings—whether the individual agents’ objectives are aligned or not. My work blends machine learning, optimization, and algorithmic game theory to create:

· communication‑efficient collaborative learning protocols that reduce sample and bandwidth costs,

· incentive‑compatible mechanisms that align self‑interested agents with system goals, and

· hierarchical decision architectures augmented with large language models, leveraging common-sense knowledge to interpret unstructured inputs and adapt to novel scenarios.

These methods enable robust, adaptive cooperative decision-making in multi-agent systems, such as intelligent industrial systems and recommender systems, ensuring sample and communication efficiency, as well as incentive awareness, under unstructured conditions.

（Full publication list: https://www.chuanhao-li.com/publications/）

1. STRIDE: A Tool-Assisted LLM Agent Framework for Strategic and Interactive Decision-Making

Chuanhao Li, Runhan Yang, Tiankai Li, Milad Bafarassat, Kourosh Sharifi, Dirk Bergemann, and Zhuoran Yang

INFORMS Workshop on Market Design @EC 2024, AutoRL Workshop @ICML 2024 [paper] [code]

2. Communication-Efficient Federated Non-Linear Bandit Optimization

Chuanhao Li*, Chong Liu*, Yu-Xiang Wang

International Conference on Learning Representations (ICLR) 2024 [paper]

3. Incentivized Truthful Communication for Federated Bandits

Zhepei Wei*, Chuanhao Li*, Haifeng Xu, Hongning Wang

International Conference on Learning Representations (ICLR) 2024 [paper]

4. PrefPaint: Aligning Image Inpainting Diffusion Model with Human Preference

Kendong Liu*, Zhiyu Zhu*, Chuanhao Li*, Hui Liu, Huanqiang Zeng, and Junhui Hou

Neural Information Processing Systems (NeurIPS) 2024 [project page]

5. Learning Kernelized Contextual Bandits in a Distributed and Asynchronous Environment

Chuanhao Li, Huazheng Wang, Mengdi Wang, Hongning Wang

International Conference on Learning Representations (ICLR) 2023 [paper]

6. Asynchronous Upper Confidence Bound Algorithms for Federated Linear Bandits

Chuanhao Li, Hongning Wang

International Conference on Artificial Intelligence and Statistics (AISTATS) 2022 [paper] [code]

7. Communication Efficient Distributed Learning for Kernelized Contextual Bandits

Chuanhao Li, Huazheng Wang, Mengdi Wang, Hongning Wang

Neural Information Processing Systems (NeurIPS) 2022 [paper]

8. Communication Efficient Federated Learning for Generalized Linear Bandits

Chuanhao Li, Hongning Wang

Neural Information Processing Systems (NeurIPS) 2022 [paper] [code]

9. When and Whom to Collaborate with in a Changing Environment: A Collaborative Dynamic Bandit Solution

Chuanhao Li, Qingyun Wu, Hongning Wang

International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) 2021 [paper] [code]

10. Unifying Clustered and Non-stationary Bandits

Chuanhao Li, Qingyun Wu, Hongning Wang

International Conference on Artificial Intelligence and Statistics (AISTATS) 2021 [paper] [code]