AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

EnvPool: A Highly Parallel Reinforcement Learning Environment Execution Engine

Jiayi Weng, Min Lin, Shengyi Huang, Bo Liu, Denys Makoviichuk, Viktor Makoviychuk, Zichen Liu, Yufan Song, Ting Luo, Yukun Jiang, Zhongwen Xu, Shuicheng YAN

NeurIPS 2022 · Reinforcement learning › Visual and representation RL · approach: Scalable RL systems and environments · 14 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Replace Python subprocess environment execution with EnvPool (async)RL environment simulation · Laptop, AtariFPS: 15,863 -> 49,439Improved
Replace Python subprocess environment execution with EnvPool (async)RL environment simulation · Laptop, MuJoCoFPS: 36,586 -> 105,126Improved
Replace Sample Factory with EnvPool (async)RL environment simulation · Atari and MuJoCo across Laptop and WorkstationFPS: 28,216 -> 49,439; 62,510 -> 105,126; 138,847 -> 200,428; 309,264 -> 582,446Improved
Use NUMA + async execution rather than Sample FactoryRL environment simulation · Atari and MuJoCo, DGX-A100FPS: 707,494 -> 1,069,922; 1,573,262 -> 3,134,287Improved
Use asynchronous rather than synchronous execution (ablation)EnvPool · Atari and MuJoCo across Laptop, Workstation, and DGX-A100FPS: 37,396 -> 49,439; 66,622 -> 105,126; 133,824 -> 200,428; 380,950 -> 582,446; 427,851 -> 891,286; 949,787 -> 2,363,864Improved
Replace Python with EnvPool for a single environmentEnvironment simulation · Atari Pong-v5, LaptopFPS: 4,891 -> 7,887Improved
Replace Python with EnvPool for a single environmentEnvironment simulation · MuJoCo Ant-v3, LaptopFPS: 12,325 -> 15,641Improved
Replace Python with EnvPool for a single environmentEnvironment simulation · DeepMind Control cheetah run, LaptopFPS: 6,235 -> 11,636Improved
Use EnvPool (Sync) instead of For-loop executionCleanRL PPO on Atari Breakout · Atari Breakout, N=8end-to-end training time: ~200 minutes -> approximately 73 minutesImproved
Use EnvPool instead of DummyVecEnvAcme PPO · MuJoCo HalfCheetah-v3wall time to similar final episode returnImproved
Use EnvPool instead of Ray parallel environmentsrl_games PPO · Atari and MuJoCo example runstraining speedImproved
Tune the number of parallel environments while keeping environment-interaction budget comparableAcme PPO with EnvPool · MuJoCo HalfCheetah-v3training timeImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy