Neural Architecture Evolution in Deep Reinforcement Learning for Continuous Control
Abstract
Current Deep Reinforcement Learning algorithms still heavily rely on handcrafted neural network architectures. We propose a novel approach to automatically find strong topologies for continuous control tasks while only adding a minor overhead in terms of interactions in the environment. To achieve this, we combine Neuroevolution techniques with off-policy training and propose a novel architecture mutation operator. Experiments on five continuous control benchmarks show that the proposed Actor-Critic Neuroevolution algorithm often outperforms the strong Actor-Critic baseline and is capable of automatically finding topologies in a sample-efficient manner which would otherwise have to be found by expensive architecture search.
Cite
@article{arxiv.1910.12824,
title = {Neural Architecture Evolution in Deep Reinforcement Learning for Continuous Control},
author = {Jörg K. H. Franke and Gregor Köhler and Noor Awad and Frank Hutter},
journal= {arXiv preprint arXiv:1910.12824},
year = {2020}
}
Comments
NeurIPS 2019 MetaLearn Workshop