Assistant Professor
The School of Artificial Intelligence (SAI)
The Chinese University of Hong Kong, Shenzhen
Google Scholar
Email: siyuanxu26@cuhk.edu.cn
Research Interests
My research focuses on the foundations and applications of reinforcement learning, with an emphasis on adaptive, safe, and self-improving intelligent systems. My research explores how agents can learn from reward models, self-generated data, and interactions with diverse environments, enabling reliable adaptation, continual improvement, and effective decision-making in complex real-world settings.
- RL theory and algorithm design: Policy optimization/adaptation, meta-RL across diverse tasks and environments, and safe RL
- RL for large model training: Reward modeling, self-distillation, and test-time training
- AI agents & self-improving AI: AI agents capable of autonomous reflection, self-improvement, and adaptation, including automatic training data generation and environment construction
- Embodied AI & robot learning: RL for robotics, robot foundation models, environment adaptation, and AI agents for robotics
Recruitment
Ph.D./MPhil/RA student recruitment: I am looking for self-motivated Ph.D. and MPhil students and research assistants (onsite and remotely) with a solid mathematical background and strong application skills. Should you be interested, please send me an email with your CV, transcript, publications (if applicable), and research interests.
I am open to all kinds of collaboration. Please email me if you are interested in my research.