SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2011.12719 · NeurIPS · 2021

RLlib Flow: Distributed Reinforcement Learning is a Dataflow Problem

U Berkeley, Joseph Gonzalez, Zhanghao Wu, Eric Liang, Michael Luo, Sven Anyscale

arXiv · PDF · Open in the Atlas

Code that ran

We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Researchers and practitioners in the field of reinforcement learning (RL) frequently leverage parallel computation, which has led to a plethora of new algorithms and systems in the last few years. In this paper, we re-examine the challenges posed by distributed RL and try to view it through the lens of an old idea: distributed dataflow. We show that viewing RL as a dataflow problem leads to highly composable and performant implementations. We propose RLlib Flow, a hybrid actor-dataflow programming model for distributed RL, and validate its practicality by porting the full suite of algorithms in RLlib, a widely adopted distributed RL library. Concretely, RLlib Flow provides 2-9× code savings in real production code and enables the composition of multi-agent algorithms not possible by end users before. The open-source code is available as part of RLlib at https://github.com/ ray-project/ray/tree/master/rllib. * indicates equal contributions.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2011.12719")
get_code_for_paper("2011.12719")
have("2011.12719")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2011.12719.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2011.12719.svg)](https://syntology.ai/paper/2011.12719)

Connect an agent — have() is free.