期刊
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
卷 34, 期 3, 页码 1454-1464出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TNNLS.2021.3105407
关键词
Task analysis; Training; Trajectory; Learning systems; Heuristic algorithms; Feature extraction; Benchmark testing; Meta-learning; reinforcement learning (RL); task adaptiveness
This article introduces an off-policy meta-RL algorithm that allows meta-learners to adapt their exploration strategy and balance task-agnostic and task-related information through latent context reorganization.
Deep reinforcement learning is confronted with problems of sampling inefficiency and poor task migration capability. Meta-reinforcement learning (meta-RL) enables meta-learners to utilize the task-solving skills trained on similar tasks and quickly adapt to new tasks. However, meta-RL methods lack enough queries toward the relationship between task-agnostic exploitation of data and task-related knowledge introduced by latent context, limiting their effectiveness and generalization ability. In this article, we develop an algorithm for off-policy meta-RL that can provide the meta-learners with self-oriented cognition toward how they adapt to the family of tasks. In our approach, we perform dynamic task-adaptiveness distillation to describe how the meta-learners adjust the exploration strategy in the meta-training process. Our approach also enables the meta-learners to balance the influence of task-agnostic self-oriented adaption and task-related information through latent context reorganization. In our experiments, our method achieves 10%-20% higher asymptotic reward than probabilistic embeddings for actor-critic RL (PEARL).
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据