I am a Postdoctoral Scholar at Stanford Vision and Learning Lab (SVL), working with Fei-Fei Li and Jiajun Wu. I received my PhD from Computer Science and Artificial Intelligence Laboratory (CSAIL) at MIT, advised by Antonio Torralba and Russ Tedrake, and my bachelor's degree from Peking University.

My research interests lie in robotics, computer vision, and machine learning. In particular, I am interested in enabling robots and embodied agents to better perceive and interact with the physical world.

Email: liyunzhu [at] stanford (dot) edu
Google Scholar / GitHub / Twitter


Recent Talk Recordings

Learning Structured World Models From and For Physical Interactions
PhD Thesis Defense at MIT CSAIL
[YouTube] (June 2022)

Graph-Structured Networks for Physical Inference and Model-Based Control
Invited talk at CVPR 2021 Tutorial Learning Representations via Graph-structured Networks
[YouTube] (June 2021)

Learning Compositional Dynamics Models for Physical Inference and Model-Based Control
Invited talk at ICLR 2021 Workshop Deep Learning for Simulation
[YouTube] (May 2021)

Publications (show selected / show by date / show by topic)
Research Topics: Robotic Manipulation / Physical Scene Understanding / Multi-Modal Perception / Imitation Learning



Selected Honors

Professional Service