MENTOR: タスク指向摂動を備えた視覚強化学習のためのMixture-of-Expertsネットワーク
MENTOR: Mixture-of-Experts Network with Task-Oriented Perturbation for Visual Reinforcement Learning
視覚強化学習のサンプル効率を改善するため、MLPをMixture-of-Expertsバックボーンに置き換え、タスク指向の摂動機構を導入した手法MENTORを提案。実機のマニピュレーションタスクで平均83%の成功率を達成した。
著者: Suning Huang, Zheyu Zhang, Tianhai Liang, Yihan Xu, Zhehao Kou, Chenhao Lu, Guowei Xu, Zhengrong Xue, Huazhe Xu
分類: cs.RO, cs.LG
原文アブストラクト
Visual deep reinforcement learning (RL) enables robots to acquire skills from visual input for unstructured tasks. However, current algorithms suffer from low sample efficiency, limiting their practical applicability. In this work, we present MENTOR, a method that improves both the architecture and optimization of RL agents. Specifically, MENTOR replaces the standard multi-layer perceptron (MLP) with a mixture-of-experts (MoE) backbone and introduces a task-oriented perturbation mechanism. MENTOR outperforms state-of-the-art methods across three simulation benchmarks and achieves an average of 83% success rate on three challenging real-world robotic manipulation tasks, significantly surpassing the 32% success rate of the strongest existing model-free visual RL algorithm. These results underscore the importance of sample efficiency in advancing visual RL for real-world robotics. Experimental videos are available at https://suninghuang19.github.io/mentor_page/.
