SURREAL-System: Fully-Integrated Stack for Distributed Deep Reinforcement Learning

TL;DR

SURREAL-System is a comprehensive, scalable framework for distributed reinforcement learning that simplifies deployment across diverse hardware and achieves state-of-the-art performance on standard benchmarks.

Contribution

It introduces a fully-integrated stack for distributed RL, combining hardware abstraction, high-level scheduling, optimized communication, and scalable algorithms.

Findings

Achieves state-of-the-art results on OpenAI Gym tasks.

Supports scaling to thousands of CPU cores and hundreds of GPUs.

Provides a flexible framework adaptable to various hardware environments.

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.