Ensemble Reinforcement Learning in Continuous Spaces -- A Hierarchical Multi-Step Approach for Policy Training

TL;DR

This paper introduces a hierarchical ensemble reinforcement learning method with multi-step training that enhances exploration, stability, and performance in continuous control tasks by promoting collaboration among base learners.

Contribution

It presents a novel multi-step integration training technique for ensemble DRL, enabling effective inter-learner collaboration and improved learning stability.

Findings

Outperforms state-of-the-art DRL algorithms on benchmark tasks

Enhances exploration and stability in continuous control environments

Theoretically verified hierarchical ensemble learning algorithm

Abstract

Actor-critic deep reinforcement learning (DRL) algorithms have recently achieved prominent success in tackling various challenging reinforcement learning (RL) problems, particularly complex control tasks with high-dimensional continuous state and action spaces. Nevertheless, existing research showed that actor-critic DRL algorithms often failed to explore their learning environments effectively, resulting in limited learning stability and performance. To address this limitation, several ensemble DRL algorithms have been proposed lately to boost exploration and stabilize the learning process. However, most of existing ensemble algorithms do not explicitly train all base learners towards jointly optimizing the performance of the ensemble. In this paper, we propose a new technique to train an ensemble of base learners based on an innovative multi-step integration method. This training technique enables us to develop a new hierarchical learning algorithm for ensemble DRL that effectively promotes inter-learner collaboration through stable inter-learner parameter sharing. The design of our new algorithm is verified theoretically. The algorithm is also shown empirically to outperform several state-of-the-art DRL algorithms on multiple benchmark RL problems.