What Matters for On-Policy Deep Actor-Critic Methods? A Large-Scale Study
Marcin Andrychowicz, Anton Raichuk, Piotr Stańczyk, Manu Orsini, Sertan Girgin, Raphaël Marinier, Leonard Hussenot, Matthieu Geist
Abstract
In recent years, reinforcement learning (RL) has been successfully applied to many different continuous control tasks. While RL algorithms are often conceptually simple, their state-of-the-art implementations take numerous low- and high-level design decisions that strongly affect the performance of the resulting agents. Those choices are usually not extensively discussed in the literature, leading to discrepancy between published descriptions of algorithms and their implementations. This makes it hard to attribute progress in RL and slows down overall progress [Engstrom'20]. As a step towards filling that gap, we implement >50 such ``"choices" in a unified on-policy deep actor-critic framework, allowing us to investigate their impact in a large-scale empirical study. We train over 250'000 agents in five continuous control environments of different complexity and provide insights and practical recommendations for the training of on-policy deep actor-critic RL agents.
BibTeX
@inproceedings{
andrychowicz2021what,
title={What Matters for On-Policy Deep Actor-Critic Methods? A Large-Scale Study},
author={Marcin Andrychowicz and Anton Raichuk and Piotr Sta{\'n}czyk and Manu Orsini and Sertan Girgin and Rapha{\"e}l Marinier and Leonard Hussenot and Matthieu Geist and Olivier Pietquin and Marcin Michalski and Sylvain Gelly and Olivier Bachem},
booktitle={International Conference on Learning Representations},
year={2021},
url={https://openreview.net/forum?id=nIAxjsniDzg}
}