Learning from Symmetry: Meta-Reinforcement Learning with Symmetrical Behaviors and Language Instructions

TL;DR

This paper introduces a dual-MDP meta-reinforcement learning approach that leverages symmetry in behaviors and language instructions to enhance generalization and learning efficiency across multiple manipulation tasks.

Contribution

It proposes a novel method combining symmetry and language instructions in meta-RL, improving task generalization and learning speed.

Findings

Significant improvement in generalization across tasks

Enhanced learning efficiency demonstrated in manipulation tasks

Method outperforms existing meta-RL approaches

Abstract

Meta-reinforcement learning (meta-RL) is a promising approach that enables the agent to learn new tasks quickly. However, most meta-RL algorithms show poor generalization in multi-task scenarios due to the insufficient task information provided only by rewards. Language-conditioned meta-RL improves the generalization capability by matching language instructions with the agent's behaviors. While both behaviors and language instructions have symmetry, which can speed up human learning of new knowledge. Thus, combining symmetry and language instructions into meta-RL can help improve the algorithm's generalization and learning efficiency. We propose a dual-MDP meta-reinforcement learning method that enables learning new tasks efficiently with symmetrical behaviors and language instructions. We evaluate our method in multiple challenging manipulation tasks, and experimental results show that our method can greatly improve the generalization and learning efficiency of meta-reinforcement learning. Videos are available at https://tumi6robot.wixsite.com/symmetry/.