Abstract
Meta-Reinforcement Learning (Meta-RL) aims to reveal shared characteristics in dynamics and reward functions across diverse training tasks. This objective is achieved by meta-learning a policy that is conditioned on task representations with encoded trajectory data or context, thus allowing rapid adaptation to new tasks from a known task distribution. However, since the trajectory data generated by the policy may be biased, the task inference module tends to form spurious correlations between trajectory data and specifc tasks, thereby leading to poor adaptation to new tasks. To address this issue, we propose the Meta-RL with task unCertAinty feedback through decoupled context-aware Reward and Dynamics components (MetaCARD). MetaCARD distinctly decouples the dynamics and rewards when inferring tasks and learning the policy, and integrates task uncertainty feedback from policy evaluation into the task inference module. This design effectively reduces uncertainty in tasks with changes in dynamics or/and reward functions, thereby enabling accurate task identifcation and adaptation. The experiment results on both Meta-World and classical MuJoCo benchmarks show that MetaCARD signifcantly outperforms prevailing MetaRL baselines, demonstrating its remarkable adaptation ability in sophisticated environments that involve changes in both reward functions and dynamics.