Effective Reinforcement Learning Control using Conservative Soft Actor-Critic

TL;DR

This paper introduces Conservative Soft Actor-Critic (CSAC), an RL algorithm that enhances stability and sample efficiency by combining entropy and relative entropy regularization, demonstrated on benchmarks and robotic simulations.

Contribution

The paper presents CSAC, a novel RL algorithm that integrates entropy and relative entropy regularization within the Actor-Critic framework for improved control performance.

Findings

CSAC outperforms existing methods in stability and efficiency.

CSAC demonstrates robustness in real-world robotic simulations.

Enhanced exploration capabilities with controlled policy updates.

Abstract

Reinforcement Learning (RL) has shown great potential in complex control tasks, particularly when combined with deep neural networks within the Actor-Critic (AC) framework. However, in practical applications, balancing exploration, learning stability, and sample efficiency remains a significant challenge. Traditional methods such as Soft Actor-Critic (SAC) and Proximal Policy Optimization (PPO) address these issues by incorporating entropy or relative entropy regularization, but often face problems of instability and low sample efficiency. In this paper, we propose the Conservative Soft Actor-Critic (CSAC) algorithm, which seamlessly integrates entropy and relative entropy regularization within the AC framework. CSAC improves exploration through entropy regularization while avoiding overly aggressive policy updates with the use of relative entropy regularization. Evaluations on benchmark tasks and real-world robotic simulations demonstrate that CSAC offers significant improvements in stability and efficiency over existing methods. These findings suggest that CSAC provides strong robustness and application potential in control tasks under dynamic environments.