Peilin Wu 吴沛琳

Hello! I am an undergraduate student of the IEEE Honor Class in Shanghai Jiao Tong University (SJTU), majoring in Computer Science and Technology. I am currently a research intern at Harvard, supervised by Prof. Yilun Du, focusing on world models for robotics, with an emphasis on multimodality, real-world adaptation, and embodiments. I'm also a member of Apex Data & Knowledge Management Lab, advised by Prof. Weinan Zhang. Previously, I'm a research intern at Shanghai AI Lab working on foundation models and humanoid robots.

My general research interests lie at the intersection of robotics and machine learning. My previous work has focused on implementing learning-based techniques to address policy adaptation challenges in quadruped locomotion. For future research, I am particularly interested in general robotics topics, including:

• How can robots effectively leverage their embodiment and integrate information from diverse modalities?
• What are the most effective strategies for integrating and benefiting from foundation models in robotic systems?
• How can robots achieve robust adaptation to varied tasks and environments through advanced data collection and generation methodologies?

CV  /  Email  /  GitHub  /  Google Scholar  

Publications

I'm interested in robotics, reinforcement learning and deep learning.

LoopSR: Looping Sim-and-Real for Lifelong Policy Adaptation of Legged Robots

This project looked into the lifelong policy adaptation situation. The idea behind was that attention should be paid to leveraging real-world data to fine-tune the policy continuously. The work proposed a pipeline to loop simulated training and real-world data collection, with only a limited amount of data to yield eminent performance in both sim-to-sim and sim-to-real experiments.

Bridging the Sim-to-Real Gap from the Information Bottleneck Perspective

This paper focused on modeling the privileged knowledge distillation problem from a theory-based perspective, where we provided mathematical analysis and a simple but effective framework HIB for the problem. Empirical experiments on both simulated and real-world tasks demonstrate that HIB yields improved generalizability compared to previous methods, which achieved about a 10% performance boost compared with baselines in various tasks.

Services

Design and source code from Jon Barron's website