Multi-Group Equivariant Augmentation for Reinforcement Learning in Robot Manipulation
This work addresses sampling efficiency for robotic manipulation, presenting an incremental improvement over prior methods limited to isometric symmetries.
The paper tackled the problem of sampling efficiency in visuomotor learning for robotic manipulation by exploring non-isometric symmetries, introducing Multi-Group Equivariance Augmentation (MEA) integrated with offline reinforcement learning, and demonstrating effectiveness through simulation and real-robot experiments.
Sampling efficiency is critical for deploying visuomotor learning in real-world robotic manipulation. While task symmetry has emerged as a promising inductive bias to improve efficiency, most prior work is limited to isometric symmetries -- applying the same group transformation to all task objects across all timesteps. In this work, we explore non-isometric symmetries, applying multiple independent group transformations across spatial and temporal dimensions to relax these constraints. We introduce a novel formulation of the partially observable Markov decision process (POMDP) that incorporates the non-isometric symmetry structures, and propose a simple yet effective data augmentation method, Multi-Group Equivariance Augmentation (MEA). We integrate MEA with offline reinforcement learning to enhance sampling efficiency, and introduce a voxel-based visual representation that preserves translational equivariance. Extensive simulation and real-robot experiments across two manipulation domains demonstrate the effectiveness of our approach.