Sample-Efficient Covariance Matrix Adaptation Evolutional Strategy via Simulated Rollouts in Neural Networks

Abstract

Gradient-free reinforcement learning algorithms often fail to scale to high dimensions and require a large number of rollouts. In this paper, we propose learning a predictor model that allows simulated rollouts in a rank-based black-box optimizer Covariance Matrix Adaptation Evolutional Strategy (CMA-ES) to achieve higher sample-efficiency. We validated the performance of our new approach on different benchmark functions where our algorithm shows a faster convergence compared to the standard CMA-ES. As a next step, we will evaluate our new algorithm in a robot cup flipping task

Publication
In 2nd International Conference on Advances in Signal Processing and Artificial Intelligence
Honghu Xue
Honghu Xue
PhD student

My research interests include Dep Reinforcement Learning and Deep Learning.