Scholar
Sizhe Wei
Google Scholar ID: KPFTRckAAAAJ
Georgia Institute of Technology
Citations & Impact
Citations
59
H-index
1
i10-index
1
Publications
4
Co-authors
12
list available
Publications
Resume (English only)
Academic Achievements
- Publications: 'Morphological-Symmetry-Equivariant Policy for Legged Robot Locomotion' and 'Morphological-Symmetry-Equivariant Heterogeneous Graph Neural Network for Robotic Dynamics Learning'; Awards: NSF Travel Grant to attend L4DC 2025.
Research Experience
- Research work during Ph.D. includes MS-PPO, a morphology-symmetry-equivariant policy learning framework, and MS-HGNN, a morphological-symmetry-equivariant heterogeneous graph neural network for robotic dynamics learning. Involved in multiple robotics projects such as Vision-Language-Action Models for Robotics and Physics-Informed Learning for Legged Locomotion.
Education
- Ph.D. Student in Robotics at Georgia Institute of Technology, advised by Prof. Lu Gan; B.Eng. and M.S. degrees in Information Engineering from Shanghai Jiao Tong University (SJTU) in 2021 & 2024, respectively, under the supervision of Prof. Ya Zhang and Prof. Siheng Chen.
Background
- Research Interests: Physics-Informed Learning for Robotics; Field: Robotics; Bio: My name, Sizhe (思哲), means 'Thinking Philosophy' in Chinese and is pronounced roughly as 'Suh-Juh'. My research focuses on integrating physical priors—such as morphological symmetry—into learning architectures to improve sample efficiency, accelerate training, and enhance generalization.
Miscellany
- Looking for motivated undergraduates and master's students to join projects in Vision-Language-Action Models for Robotics and Physics-Informed Learning for Legged Locomotion.
