# DeepMind의 인공지능 AlphaStar, 스타크래프트2 그랜드마스터레벨에 등극

> Clean Markdown view of GeekNews topic #828. Use the original source for factual precision when an external source URL is present.

## Metadata

- GeekNews HTML: [https://news.hada.io/topic?id=828](https://news.hada.io/topic?id=828)
- GeekNews Markdown: [https://news.hada.io/topic/828.md](https://news.hada.io/topic/828.md)
- Type: news
- Author: [xguru](https://news.hada.io/@xguru)
- Published: 2019-11-01T08:35:30+09:00
- Updated: 2019-11-01T08:35:30+09:00
- Original source: [deepmind.com](https://deepmind.com/blog/article/AlphaStar-Grandmaster-level-in-StarCraft-II-using-multi-agent-reinforcement-learning)
- Points: 5
- Comments: 1

## Topic Body

"멀티 에이전트 강화학습" 훈련을 한 AI가 처음으로 GM(상위 0.2%)레벨 에 오름.

논문과 경기 다시보기도 제공

## Comments



### Comment 569

- Author: xguru
- Created: 2019-11-01T08:37:50+09:00
- Points: 1

경기 Replay : https://deepmind.com/research/open-source/alphastar-resources

Nature 기사에는 관련된 자료 링크들이 잘 정리되어 있음 : https://www.nature.com/articles/s41586-019-1724-z