Select language
< Return to main menu

Huazhe Xu

PI(September 2022 to present)
Assistant Professor


XU Huazhe is currently a tenure-track Assistant Professor at the Institute for Interdisciplinary Information Sciences, Tsinghua University. He completed his postdoctoral work at Stanford University and earned his Ph.D. from the University of California, Berkeley. In 2021, he served as a part-time visiting scientist in the Meta Artificial Intelligence research department. His research focuses on the theory, algorithms, and applications of embodied artificial intelligence (Embodied AI), including deep reinforcement learning, robotics, and sensorimotor control. He systematically studies visual deep reinforcement learning, imitation learning, and robot manipulation, with the aim of addressing core issues such as low data efficiency and weak generalization ability in embodied artificial intelligence. He has published over 30 top-tier conference papers, and his representative works have been covered by media outlets such as MIT Tech Review and Stanford HAI.

Research Direction

Deep reinforcement learning 


Computer tactile sensing

Computer vision

Research topic

AI- Open world embodied artificial intelligence



Open positions

Research Direction:

Embodied Intelligence: Robotics, Reinforcement Learning, Tactile Sensing, 3D Computer Vision


1. Responsible for research, algorithm development or system development in the aforementioned related fields and directions; 

2. Publish academic or innovative research achievements in the aforementioned related fields.

Required Qualification:

Major in computer science, electronics, automation, software engineering, physics or related fields, with strong academic ability; 

Possess excellent theoretical knowledge and programming skills (Python, Linux, C++, etc.) in the relevant field; 

Strong self-motivation, curious about the world, and enjoy exploring new things.

Please send your CV:


One paper is accepted by ICML'23.

I will serve as an SPC member in IJCAI'23.

Four papers are accepted by ICLR'23.

Four papers are accepted by ICRA'23.

Two papers are accepted by NeurIPS'22.

One paper is accepted by CORL'22.

Our paper "Scene Synthesis from Human Motion" is accepted by Siggraph Asia'22.

Our paper "Towards Learning to Play Piano with Dexterous Hands and Touch" is accepted by IROS'22.

Our RSS paper "RoboCraft" is covered by MIT Tech Review and HAI.

Two papers are accepted by ICML'22.

Our paper "Don't Touch What Matters" is accepted by IJCAI'22.

I gave a talk at UC San Diego.

Our paper RoboCraft is accepted by RSS'22.

One paper is accepted by ICLR'22 (Spotlight).

Two papers are accepted by Neurips'21.


Tianying Ji, Yu Luo, Fuchun Sun, Xianyuan Zhan, Jianwei Zhang, Huazhe Xu. Seizing Serendipity: Exploiting the Value of Past Success in Off-Policy Actor-CriticarXiv preprint, 2023.

Nicklas Hansen*, Zhecheng Yuan*, Yanjie Ze*, Tongzhou Mu*, Aravind Rajeswaran+, Hao Su+, Huazhe Xu+, Xiaolong Wang+. On Pre-Training for Visuo-Motor Control: Revisiting a Learning-from-Scratch Baseline International Conference on Machine Learning (ICML), 2023.

Zhengrong Xue, Zhecheng Yuan, Jiashun Wang, Xueqian Wang, Yang Gao, Huazhe Xu. USEEK: Unsupervised SE(3)-Equivariant 3D Keypoints for Generalizable Manipulation International Conference on Robot Automation (ICRA), 2023.

Changyi Lin, Ziqi Lin, Shaoxiong Wang, Huazhe Xu. DTact: A Vision-Based Tactile Sensor that Measures High-Resolution 3D Geometry Directly from Darkness International Conference on Robot Automation (ICRA), 2023.

Ray Chen Zheng*, Kaizhe Hu*, Zhecheng Yuan, Boyuan Chen, Huazhe Xu. Extraneousness-Aware Imitation Learning International Conference on Robot Automation (ICRA), 2023.

Yunfei Li*, Chaoyi Pan*, Huazhe Xu, Xiaolong Wang, Yi Wu. Efficient Bimanual Handover and Rearrangement via Symmetry-Aware Actor-Critic Learning International Conference on Robot Automation (ICRA), 2023.

Kaizhe Hu*, Ray Zheng*, Yang Gao, Huazhe Xu. Decision Transformer under Random Frame Dropping International Conference on Learning Representation (ICLR), 2023.

Pu Hua, Yubei Chen+, Huazhe Xu+. Simple Emergent Action Representations from Multi-Task Policy Training International Conference on Learning Representation (ICLR), 2023.

Linfeng Zhao, Huazhe Xu, Lawson L.S. Wong. Scaling up and Stabilizing Differentiable Planning with Implicit Differentiation International Conference on Learning Representation (ICLR), 2023.