TorchBeast: A PyTorch Platform for Distributed RL
Published in
arXiv.org(2019)External Links:
TL
TL;DR
The TorchBeast design principles and implementation are described and it is demonstrated that it performs on-par with IMPALA on Atari.
Abstract
TorchBeast is a platform for reinforcement learning (RL) research in PyTorch. It implements a version of the popular IMPALA algorithm for fast, asynchronous, parallel training of RL agents. Additionally, TorchBeast has simplicity as an explicit design goal: We provide both a pure-Python implementation ("MonoBeast") as well as a multi-machine high-performance version ("PolyBeast"). In the latter, parts of the implementation are written in C++, but all parts pertaining to machine learning are kept in simple Python using PyTorch, with the environments provided using the OpenAI Gym interface. This enables researchers to conduct scalable RL research using TorchBeast without any programming knowledge beyond Python and PyTorch. In this paper, we describe the TorchBeast design principles and implementation and demonstrate that it performs on-par with IMPALA on Atari. TorchBeast is released as an open-source package under the Apache 2.0 license and is available at \url{this https URL}.
References22 items
1
MVFST-RL: An Asynchronous RL Framework for Congestion Control with Delayed Actions
2
MinAtar: An Atari-Inspired Testbed for Thorough and Reproducible Reinforcement Learning Experiments
3
A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play
4
High-Level Strategy Selection under Partial Observability in StarCraft: Brood War
5