👤 About me

I am a Ph.D. student at the Institute for Artificial Intelligence, Peking University. I am working under the supervision of Prof. Yaodong Yang, my first initiation mentor in multi-agent systems. My research interests include reinforcement learning, game theory, and their intersection with fundamental science. The current topics I am focused on are:

• Multi-agent Learning: Developing practical algorithms for large-scale system control and multi-player computer games.
• Reinforcement Learning for Science: Exploring the potential of reinforcement learning in fundamental scientific problems.

🔥 News

• 2025.01:  🎉🎉 Two papers get accepted in ICLR 2025.
• 2024.12:  🎉🎉 One paper gets accepted in AAMAS 2025.
• 2024.12:  🎉🎉 One paper gets accepted in AAAI 2025.
• 2024.09:  🎉🎉 One paper gets accepted in NeurIPS 2024.
• 2024.07:  🎉🎉 One paper gets accepted on Nature Machine Intelligence.

📝 Publications

Nature Machine Intelligence

Efficient and Scalable Reinforcement Learning for Large-scale Network Control

Chengdong Ma*, Aming Li*, Yali Du*, Hao Dong, Yaodong Yang

This work has been covered by:
新华网 XinHua Net
科技日报 Science and Technology Daily
北京大学新闻网 Peking University News

ICLR 2025

Magnetic Preference Optimization: Achieving Last-iterate Convergence for Language Model Alignment

Mingzhi Wang, Chengdong Ma, Qizhi Chen, Linjian Meng, Yang Han, Jiancong Xiao, Zhaowei Zhang, Jing Huo, Weijie J. Su, Yaodong Yang

AAMAS 2025

Mean Field Correlated Imitation Learning

Zhiyu Zhao, Chengdong Ma, Qirui Mi, Ning Yang, Xue Yan, Mengyue Yang, Haifeng Zhang, Jun Wang, Yaodong Yang

NeurIPS 2024

Panacea: Pareto Alignment via Preference Adaptation for LLMs

Yifan Zhong*, Chengdong Ma*, Xiaoyuan Zhang*, Ziran Yang, Haojun Chen, Qingfu Zhang, Siyuan Qi, Yaodong Yang

Arxiv 2023

Evolving Diverse Red-team Language Models in Multi-round Multi-agent Games

Chengdong Ma*, Ziran Yang*, Hai Ci, Jun Gao, Minquan Gao, Xuehai Pan, Yaodong Yang

IROS 2022

Scalable model-based policy optimization for decentralized networked systems

Yali Du*, Chengdong Ma*, Yuchen Liu, Runji Lin, Hao Dong, Jun Wang, Yaodong Yang

IEEE Access

Confrontation and Obstacle-Avoidance of Unmanned Vehicles Based on Progressive Reinforcement Learning

Chengdong Ma, Jianan Liu, Saichao He, Wenjing Hong, Jia Shi

📦 Resources

ICML Tutorial

Alignment Methods in Large Language Models

Mingzhi Wang, Chengdong Ma, Yaodong Yang

🎙 Talks

• 2024.12: Invited Talk in Distributed Artificial Intelligence (DAI) conference.
• 2024.10: Invited Talk at National Key Lab of Autonomous Intelligent Unmanned Systems in Beijing Institute of Technology.
• 2024.09: Invited Talk at Cognitive Computing and Reasoning Lab in Beijing Institute for General Artificial Intelligence.