profile photo

Tianyu Li - 黎天宇

I am a CS Phd at Georgia Tech working with Prof.Sehoon Ha. I was a research scientist at Meta(Facebook) AI Research.

My research interests generally lie at the intersection of Reinforcement Learning, Robotics and Computer Animation. My current focus is on transferring human motion to robots with significantly different morphology. This technique will allow any robot learning skills directly from human demonstrations. Beside this direction, I'm also working on several other projects related to agile locomotion and reinforcement learning.

Google Scholar  /  Twitter

profile photo
  • One paper submitted to SIGGRAPH 2023
  • One paper submitted to RSS 2023
Selected Research
PontTuset ACE: Adversarial Correspondence Embedding for Cross Morphology Motion Retargeting from Human to Robot
Tianyu Li, Jungdam Won, Alexander Clegg, Akshara Rai, Sehoon Ha
Submitted to SIGGRAPH, 2023
Project Page / Paper / Video

PontTuset FastMimic: Model-based Motion Imitation for Agile, Diverse and Generalizable Quadrupedal Locomotion
Tianyu Li, Jungdam Won, Sehoon Ha, Akshara Rai
Submitted to RA-L, 2022
Paper / Video

PontTuset Planning in learned latent action spaces for generalizable legged locomotion
Tianyu Li, Roberto Calandra, Deepak Pathak, Yuandong Tian, Franziska Meier, Akshara Rai
RA-L, 2021
Paper / Video

PontTuset Learning generalizable locomotion skills with hierarchical reinforcement learning
Tianyu Li, Nathan Lambert, Roberto Calandra, Franziska Meier, Akshara Rai
ICRA, 2020
Paper / Video

PontTuset Using deep reinforcement learning to learn high-level policies on the atrias biped
Tianyu Li, Hartmut Geyer, Christopher G Atkeson, Akshara Rai
ICRA, 2019

Modified version of template from this.