Ruizhe Liu

I’m a PhD candidate at the HKU Musketeers Foundation Institute of Data Science, The University of Hong Kong, where I work under the supervision of Prof. Yanchao Yang and Prof. Yi Ma. My research lies at the intersection of machine learning and robotics, with a particular focus on enabling robots to learn complex manipulation skills from demonstrations.

🔬 Research Interests

Do you think that natural language is the most suitable language for robots?

I believe robots need their own representation system—one that is naturally suited for learning skills and sharing knowledge, rather than relying solely on human-centric abstractions. My research explores how robots can develop this autonomous knowledge system for increasingly intelligent behavior.

The Core Vision: Enable robots to discover, represent, and share knowledge in ways that are grounded in their sensorimotor experience, building structures that support efficient learning and knowledge transfer across tasks.

My Research Focuses On:

• Concept Learning: Automatical discovery of task-relevant concepts from unlabeled demonstrations, learning representations grounded in their own sensorimotor experience

• Hierarchical Learning: Robots learn manipulation primitives and compose them into complex skills, creating structured knowledge systems for skill acquisition and sharing

• Robust Manipulation: Systems that generalize across unstructured environments, ensuring learned representations remain useful and transferable

🎓 Education

PhD in Data Science (2024 - 2028 Expected)
The University of Hong Kong, Hong Kong SAR

• Funded by the Hong Kong PhD Fellowship Scheme
• Also recognized as HKU Presidential PhD Scholar
  
  

Bachelor of Science in Information and Computing Science (2020 - 2024)
Peking University, Beijing, China