Equivariant Policy Learning for Robotic Manipulation
Abstract: Despite recent advances in machine learning for robotics, current approaches often lack sample efficiency, posing a significant challenge due to the enormous time required to collect real-robot data. In this talk, I will present our innovative methods that tackle this challenge by leveraging the inherent symmetries in the physical environment. Specifically, I will outline a comprehensive framework of equivariant policy learning and its application across various robotic problem settings, including reinforcement learning, behavior cloning, and grasping. Our methods significantly outperform state-of-the-art baselines while achieving these results with far less data, both in simulation and real-world scenarios. Furthermore, our approach demonstrates robustness in the presence of symmetry distortions, such as variations in camera angles.
Bio: Dian Wang is a Ph.D. candidate at the Khoury College of Computer Sciences, Northeastern University, where he is co-advised by Prof. Robert Platt and Prof. Robin Walters. His research lies at the intersection of Machine Learning and Robotics, with a focus on Geometric Deep Learning and its applications in Robot Learning. Recently, Dian has focused on improving robotic manipulation through the use of equivariant methods to boost learning efficiency and performance. He has contributed to leading conferences and journals, including ICLR, NeurIPS, CoRL, RSS, and IJRR. Dian was awarded the JPMorgan Ph.D. Fellowship in 2023 and the Khoury College Graduate Research Fellowship in 2019.