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

Honghu Xue, Sven Böttger, Nils Rottmann, Harit Pandya, Ralf Bruder, Gerdhard Neumann, Elmar Rueckert

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.
OriginalspracheEnglisch
Seitenumfang6
PublikationsstatusVeröffentlicht - 2020
Veranstaltung2nd International Conference on Advances in Signal Processing and Artificial Intelligence - Berlin, Deutschland
Dauer: 01.03.202003.03.2020

Tagung, Konferenz, Kongress

Tagung, Konferenz, Kongress2nd International Conference on Advances in Signal Processing and Artificial Intelligence
KurztitelASPAI' 2020
Land/GebietDeutschland
OrtBerlin
Zeitraum01.03.2003.03.20

Fingerprint

Untersuchen Sie die Forschungsthemen von „Sample-Efficient Covariance Matrix Adaptation Evolutional Strategy via Simulated Rollouts in Neural Networks“. Zusammen bilden sie einen einzigartigen Fingerprint.

Zitieren