Importance mixing: Improving sample reuse in evolutionary policy search methods - Archive ouverte HAL Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2019

Importance mixing: Improving sample reuse in evolutionary policy search methods

Résumé

Deep neuroevolution, that is evolutionary policy search methods based on deep neural networks, have recently emerged as a competitor to deep reinforcement learning algorithms due to their better parallelization capabilities. However, these methods still suffer from a far worse sample efficiency. In this paper we investigate whether a mechanism known as "importance mixing" can significantly improve their sample efficiency. We provide a didactic presentation of importance mixing and we explain how it can be extended to reuse more samples. Then, from an empirical comparison based on a simple benchmark, we show that, though it actually provides better sample efficiency, it is still far from the sample efficiency of deep reinforcement learning, though it is more stable.

Dates et versions

hal-02397754 , version 1 (06-12-2019)

Identifiants

Citer

Aloïs Pourchot, Nicolas Perrin, Olivier Sigaud. Importance mixing: Improving sample reuse in evolutionary policy search methods. 2019. ⟨hal-02397754⟩
54 Consultations
1 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More