Empirical Evaluation of Contextual Policy Search with a Comparison-based Surrogate Model and Active Covariance Matrix Adaptation

Alexander Fabisch

In: Proceedings of the Genetic and Evolutionary Computation Conference Companion. Genetic and Evolutionary Computation Conference (GECCO-2019) July 13-17 Prag Czech Republic 2019.


Contextual policy search (CPS) is a class of multi-task reinforcement learning algorithms that is particularly useful for robotic applications. A recent state-of-the-art method is Contextual Covariance Matrix Adaptation Evolution Strategies (C-CMA-ES). It is based on the standard black-box optimization algorithm CMA-ES. There are two useful extensions ofCMA-ES thatwe will transfer to C-CMA-ES and evaluate empirically: ACM-ES, which uses a comparison-based surrogate model, and aCMA-ES, which uses an active update of the covariance matrix. We will show that improvements with these methods can be impressive in terms of sample-efficiency, although this is not relevant any more for the robotic domain.

German Research Center for Artificial Intelligence
Deutsches Forschungszentrum für Künstliche Intelligenz