Efficient Multi-Task and Transfer Reinforcement Learning with Parameter-Compositional Framework
This work addresses the problem of sample efficiency and performance in multi-task and transfer reinforcement learning for domains like manipulation tasks, but it appears incremental as it builds on existing multi-task and transfer methods.
The paper tackles the challenge of improving multi-task training and transfer in reinforcement learning by proposing a parameter-compositional framework, resulting in enhanced performance in multi-task training and effective transfer with better sample efficiency and performance.
In this work, we investigate the potential of improving multi-task training and also leveraging it for transferring in the reinforcement learning setting. We identify several challenges towards this goal and propose a transferring approach with a parameter-compositional formulation. We investigate ways to improve the training of multi-task reinforcement learning which serves as the foundation for transferring. Then we conduct a number of transferring experiments on various manipulation tasks. Experimental results demonstrate that the proposed approach can have improved performance in the multi-task training stage, and further show effective transferring in terms of both sample efficiency and performance.