Yancheng Liang

CSE Ph.D. Student, University of Washington | Reinforcement Learning, Multi-agent Systems

portrait_1.jpg

Yancheng Liang

yancheng@cs.washington.edu

My research interest lies in exploring the unique intelligence that emerges from collective/social behaviors such as cooperation, adaptation, imitation, teaching, confrontation and so on. I am particularly focused on applying machine learning techniques—especially reinforcement learning—to study such multi-agent systems. This also involves leveraging tools from game theory, theoretical aspects of reinforcement learning, agent modeling, and analytical models derived from social sciences.

I am currently pursuing a Ph.D. at University of Washington, CSE, co-advised by Simon Shaolei Du and Natasha Jaques. Before that, I completed my undergraduate studies at Tsinghua University, Institute for Interdisciplinary Information Sciences (Yao Class), guided by Yi Wu.

news

Jan 27, 2026 One paper accepcted by ICLR 2026!
Jun 18, 2025 One paper accepcted by ICML 2025 as an oral paper!
Sep 25, 2024 One paper accepcted by NeurIPS 2024!
Jan 04, 2024 Join University of Washington as a CSE Ph.D. student!

selected publications

  1. Improving Human-AI Coordination through Adversarial Training and Generative Models
    Paresh Chaudhary, Yancheng Liang, Daphne Chen, Simon S Du, and Natasha Jaques
    The Fourteenth International Conference on Learning Representations, 2025
  2. Cross-environment Cooperation Enables Zero-shot Multi-agent Coordination
    Kunal Jha, Wilka Carvalho, Yancheng Liang, Simon Shaolei Du, Max Kleiman-Weiner*, and Natasha Jaques*
    Forty-second International Conference on Machine Learning (Oral), 2025
  3. Learning to Cooperate with Humans using Generative Agents
    Yancheng Liang, Daphne Chen, Abhishek Gupta, Simon Shaolei Du*, and Natasha Jaques*
    Advances in Neural Information Processing Systems, 2024, 2024