SURREAL-System: Fully-Integrated Stack for Distributed Deep Reinforcement Learning
Abstract
We present an overview of SURREAL-System, a reproducible, flexible, and scalable framework for distributed reinforcement learning (RL). The framework consists of a stack of four layers: Provisioner, Orchestrator, Protocol, and Algorithms. The Provisioner abstracts away the machine hardware and node pools across different cloud providers. The Orchestrator provides a unified interface for scheduling and deploying distributed algorithms by high-level description, which is capable of deploying to a wide range of hardware from a personal laptop to full-fledged cloud clusters. The Protocol provides network communication primitives optimized for RL. Finally, the SURREAL algorithms, such as Proximal Policy Optimization (PPO) and Evolution Strategies (ES), can easily scale to 1000s of CPU cores and 100s of GPUs. The learning performances of our distributed algorithms establish new state-of-the-art on OpenAI Gym and Robotics Suites tasks.
Cite
@article{arxiv.1909.12989,
title = {SURREAL-System: Fully-Integrated Stack for Distributed Deep Reinforcement Learning},
author = {Linxi Fan and Yuke Zhu and Jiren Zhu and Zihua Liu and Orien Zeng and Anchit Gupta and Joan Creus-Costa and Silvio Savarese and Li Fei-Fei},
journal= {arXiv preprint arXiv:1909.12989},
year = {2019}
}
Comments
Technical report of the SURREAL system. See more details at https://surreal.stanford.edu