Trust-PCL: An Off-Policy Trust Region Method for Continuous Control

TL;DR

Trust-PCL introduces an off-policy trust region method for continuous control in reinforcement learning, enhancing sample efficiency and solution quality by leveraging multi-step pathwise consistencies.

Contribution

It proposes a novel off-policy trust region algorithm, Trust-PCL, based on pathwise consistencies, improving stability and sample efficiency over traditional on-policy methods like TRPO.

Findings

Trust-PCL outperforms TRPO in continuous control tasks.

It achieves higher sample efficiency and better solution quality.

The method maintains stability while exploiting off-policy data.

Abstract

Trust region methods, such as TRPO, are often used to stabilize policy optimization algorithms in reinforcement learning (RL). While current trust region strategies are effective for continuous control, they typically require a prohibitively large amount of on-policy interaction with the environment. To address this problem, we propose an off-policy trust region method, Trust-PCL. The algorithm is the result of observing that the optimal policy and state values of a maximum reward objective with a relative-entropy regularizer satisfy a set of multi-step pathwise consistencies along any path. Thus, Trust-PCL is able to maintain optimization stability while exploiting off-policy data to improve sample efficiency. When evaluated on a number of continuous control tasks, Trust-PCL improves the solution quality and sample efficiency of TRPO.