ロバストなSim-to-Real強化学習のためのデュアルアクションポリシー
Dual Action Policy for Robust Sim-to-Real Reinforcement Learning
単一のポリシーでタスク遂行用とドメイン適応用の2種類の行動を同時に予測し、不確実性に基づく探索も取り入れることで、シミュレーションと実機のギャップを埋める強化学習手法を提案した。
著者: Ng Wen Zheng Terence, Chen Jianda
分類: cs.LG, cs.AI, cs.RO
原文アブストラクト
This paper presents Dual Action Policy (DAP), a novel approach to address the dynamics mismatch inherent in the sim-to-real gap of reinforcement learning. DAP uses a single policy to predict two sets of actions: one for maximizing task rewards in simulation and another specifically for domain adaptation via reward adjustments. This decoupling makes it easier to maximize the overall reward in the source domain during training. Additionally, DAP incorporates uncertainty-based exploration during training to enhance agent robustness. Experimental results demonstrate DAP's effectiveness in bridging the sim-to-real gap, outperforming baselines on challenging tasks in simulation, and further improvement is achieved by incorporating uncertainty estimation.