AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

High-Throughput Synchronous Deep RL

Iou-Jen Liu, Raymond Yeh, Alexander Schwing

NeurIPS 2020 · Reinforcement learning › Visual and representation RL · approach: Scalable RL systems and environments · 9 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
Use HTS-RL rather than synchronous A2C/PPOdeep RL training · Atari and GFootballtraining speed: 2-6× fasterImproved
Use HTS-RL instead of A2CAtari · Atari BankHeistaverage evaluation reward: 775 ± 166 -> 942 ± 100Improved
Use HTS-RL instead of PPOGFootball · 3 vs 1 with keeperaverage score: 0.20±0.09 -> 0.81±0.02Improved
Increase actor count from 1 to 4 (ablation)HTS-RL · GFootball 3 vs 1 with keepersteps per second: 1229 -> 1362Improved
Increase actor count from 4 to 8 or 16 (ablation)HTS-RL · GFootball 3 vs 1 with keepersteps per second: 1362 -> 1393; 1362 -> 1388No clear effect
Increase actor count from 1 to 16 (ablation)HTS-RL · GFootball 3 vs 1 with keeperaverage score: 0.82 -> 0.82No clear effect
Increase synchronization interval from 4 to 512 (ablation)HTS-RL · GFootball 3 vs 1 with keepersteps per second: 445 -> 1377Improved
Vary synchronization interval from 4 to 512 (ablation)HTS-RL · GFootball 3 vs 1 with keeperaverage score: 0.82; 0.81; 0.81; 0.82; 0.81; 0.83No clear effect
Use one-step delayed gradient instead of truncated importance sampling (ablation)HTS-RL · Atari BankHeist, Breakout, and Seaquestaverage episode reward: 987 vs 881; 415 vs 390; 1831 vs 1827Improved
Use one-step delayed gradient instead of no correction (ablation)HTS-RL · Atari BankHeist, Breakout, and Seaquestaverage episode reward: 987 vs 877; 415 vs 402; 1831 vs 1784Improved
Train 3 agents instead of 1HTS-RL (PPO), raw-image input · GFootball 3 vs 1 with keeperaverage game score: 0.30± 0.11 -> 0.63± 0.15Improved

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