Time Reversal Symmetry for Efficient Robotic Manipulations in Deep Reinforcement Learning
This work addresses sample efficiency in robotics tasks with temporal symmetries, such as door opening and closing, but it is incremental as it extends existing symmetry-based methods to a new type of symmetry.
The paper tackled the problem of improving sample efficiency in deep reinforcement learning for robotics by exploring time reversal symmetry, which had been neglected in favor of spatial symmetries. The result was the TR-DRL framework, which achieved higher sample efficiency and stronger final performance on benchmarks like Robosuite and MetaWorld compared to baselines.
Symmetry is pervasive in robotics and has been widely exploited to improve sample efficiency in deep reinforcement learning (DRL). However, existing approaches primarily focus on spatial symmetries, such as reflection, rotation, and translation, while largely neglecting temporal symmetries. To address this gap, we explore time reversal symmetry, a form of temporal symmetry commonly found in robotics tasks such as door opening and closing. We propose Time Reversal symmetry enhanced Deep Reinforcement Learning (TR-DRL), a framework that combines trajectory reversal augmentation and time reversal guided reward shaping to efficiently solve temporally symmetric tasks. Our method generates reversed transitions from fully reversible transitions, identified by a proposed dynamics-consistent filter, to augment the training data. For partially reversible transitions, we apply reward shaping to guide learning, according to successful trajectories from the reversed task. Extensive experiments on the Robosuite and MetaWorld benchmarks demonstrate that TR-DRL is effective in both single-task and multi-task settings, achieving higher sample efficiency and stronger final performance compared to baseline methods.