Maximum Entropy On-Policy Actor-Critic via Entropy Advantage Estimation

Jean Seong Bjorn Choe, Jong-Kook Kim

Keywords: policy gradient, entropy regularization, maximum entropy reinforcement learning, actor-critic

TL;DR: We propose a practical implementation of the MaxEnt RL framework that extends existing deep on-policy actor-critic algorithms like PPO, utilising an additional critic to separately evaluate the augmented entropy objective and show its efficiency.

Abstract: Entropy Regularisation is a widely adopted technique that enhances policy optimisation performance and stability. A notable form of entropy regularisation is augmenting the objective with an entropy term, thereby simultaneously optimising the expected return and the entropy. This framework, known as maximum entropy reinforcement learning (MaxEnt RL), has shown theoretical and empirical successes. However, its practical application in straightforward on-policy actor-critic settings remains surprisingly underexplored. We hypothesise that this is due to the difficulty of managing the entropy reward in practice. This paper proposes a simple method of separating the entropy objective from the MaxEnt RL objective, which facilitates the implementation of MaxEnt RL in on-policy settings. Our empirical evaluations demonstrate that extending Proximal Policy Optimisation (PPO) and Trust Region Policy Optimisation (TRPO) within the MaxEnt framework improves policy optimisation performance in both MuJoCo and Procgen tasks. Additionally, our results highlight MaxEnt RL's capacity to enhance generalisation.

Submission Number: 86