← Flash PapersDota 2 with Large Scale Deep Reinforcement Learning — OpenAI Five

Share this Flash Paper

Dota 2 with Large Scale Deep Reinforcement Learning — OpenAI Five

https://flashpapers.ai/p/a67dxsnjgb
Post to XPost to BlueskyShare on LinkedInSubmit to Hacker NewsSubmit to RedditEmail a link
Make your own
OpenAI, : +25 more
cs.LG2019arXiv:1912.06680
Details ▸
Win rate vs Public:99.4%· OpenAI Five Arena (7,257 games)Match Score vs World Champions:2-0· Team OG ShowmatchOptimization Compute:770 +/- 50 PFlops/s-daysAverage Reaction Time:217ms· Dota 2
Concepts:Proximal Policy Optimization
Dota 2 with Large Scale Deep Reinforcement Learning — OpenAI Five — Flash Papers