Research Areas
Our team at the Department of Data and Systems Engineering, The University of Hong Kong, specializes in Learning for Sequential Decision-Making and Robotic Control.
We are interested in advancing the following research areas:
- Development of Robotic Foundation Models.
- End-to-end multi-modal learning for Humanoid Robotic Control and its industrial applications.
- Post-training techniques for Large Language Model agents and their applications in complex reasoning or sequential decision-making tasks.
- Developing a unified theoretical framework for Reinforcement Learning, Optimal Control, and Stochastic Optimization.
- Investigation of scaling laws in sequential decision-making, with particular emphasis on scaling with data (e.g., data-driven decision-making) and computation (e.g., Monte Carlo Tree Search systems like AlphaZero).
Currently, we have three active research directions: Offline Reinforcement Learning for Controllable Nuclear Fusion, Continual Reinforcement Learning for Humanoid Robotic Control, and Learning-based Ego-centric Evolution within a Multi-agent System, of which more details are provided in the introduction.
News
- 2025-11-08: Two papers accepted at AAAI 2026, and one of them will be presented orally 🎉🎉🎉
- 2025-10-21: One paper accepted for the AAAI Student Abstract and Poster Program 2026, featuring an oral presentation 🎉🎉🎉
- 2025-10-09: Our RL-LLM framework Verlog is featured by JIQIZHIXIN 🌐🌐🌐
- 2025-09-19: Two papers accepted at NeurIPS 2025 🎉🎉🎉
- 2025-07-22: Our survey on Behavior Foundation Models is featured by JIQIZHIXIN 🌐🌐🌐
How to join us
Current Openings
(If interested, please contact Dr. Jiayu Chen directly. The Ph.D. application system accepts submissions year-round.)
- Fully funded Ph.D. positions starting in Fall 2025 have been taken.
- Openings for self-financed Ph.D. students and research assistants are available year-round.
- ❗❗❗ We have available positions for (funded) research assistants on World Models and Robotic Foundation Models.
- ❗❗❗ We need several remote research interns to work on skill adaptation, real-to-sim, generative-model-based RL, or auto-deployment of embodiment intelligence. Please reach out to Dr. Jiayu Chen if interested. Our Ph.D. or MPhil candidates are usually selected among research interns.
Qualifications for Ph.D. and Postdoctoral Researchers:
Candidates should possess strong proficiency in mathematics or programming, as well as demonstrate self-motivation and resilience. Applicants with backgrounds in the following areas are highly preferred: Humanoid Robots, Large Language/Reasoning Models, Control Theory, Optimization Theory, Statistical Machine Learning.
For more information, please refer to the respective categories in openings.