Efficient Multi-Task and Transfer Reinforcement Learning with Parameter-Compositional Framework

2023/06/02 by Lingfeng Sun, Sun, Lingfeng, Haichao Zhang +5 · 1 citation
Engineering · Neuroscience · Computer Science · #Muscle activation and electromyography studies #EEG and Brain-Computer Interfaces #Reinforcement Learning in Robotics

paper · pdf · doi:10.48550/arxiv.2306.01839

Abstract

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.

Cited by

• One Model for All Tasks: Leveraging Efficient World Models in Multi-Task Planning

Related

• Multi-Source Transfer Learning for Deep Model-Based Reinforcement Learning
• Transfer from Multiple MDPs
• Cross-Domain Transfer in Reinforcement Learning using Target Apprentice
• Sample Complexity of Multi-task Reinforcement Learning
• Measuring and Harnessing Transference in Multi-Task Learning