An Efficient Asynchronous Method for Integrating Evolutionary and Gradient-based Policy Search
This work aims to improve the efficiency and stability of policy search for deep reinforcement learning practitioners by integrating evolutionary and gradient-based methods asynchronously.
This paper introduces an Asynchronous Evolution Strategy-Reinforcement Learning (AES-RL) framework that combines ES and DRL to address the limitations of synchronous update schemes. The proposed method demonstrates superior performance and time efficiency on continuous control benchmarks compared to existing approaches.
Deep reinforcement learning (DRL) algorithms and evolution strategies (ES) have been applied to various tasks, showing excellent performances. These have the opposite properties, with DRL having good sample efficiency and poor stability, while ES being vice versa. Recently, there have been attempts to combine these algorithms, but these methods fully rely on synchronous update scheme, making it not ideal to maximize the benefits of the parallelism in ES. To solve this challenge, asynchronous update scheme was introduced, which is capable of good time-efficiency and diverse policy exploration. In this paper, we introduce an Asynchronous Evolution Strategy-Reinforcement Learning (AES-RL) that maximizes the parallel efficiency of ES and integrates it with policy gradient methods. Specifically, we propose 1) a novel framework to merge ES and DRL asynchronously and 2) various asynchronous update methods that can take all advantages of asynchronism, ES, and DRL, which are exploration and time efficiency, stability, and sample efficiency, respectively. The proposed framework and update methods are evaluated in continuous control benchmark work, showing superior performance as well as time efficiency compared to the previous methods.