3 ms·
Much More Efficient Self-Play Learning in Go
- jedharris 6y agoRelated demos, repositories, papers and posts: More recent and extensive post: https://blog.janestreet.com/deep-learning-the-hardest-go-problem-in-the-world/ https://blog.janestreet.com/deep-learning-the-hardest-go-pro... Site to play against KataGo: https://online-go.com/player/592684/ https://online-go.com/player/592684/ Arxiv paper on KataGo design: https://arxiv.org/abs/1902.10565 https://arxiv.org/abs/1902.10565 GitHub repository of KataGo: https://github.com/lightvector/KataGo https://github.com/lightvector/KataGo Another writer commenting on the history of Go AI, and significance of KataGo: https://brantondemoss.com/writing/kata.html https://brantondemoss.com/writing/kata.html
- jedharris 6y agoThe author's most recent version of KataGo averages about 50 times greater learning efficiency from self-play. See <https://blog.janestreet.com/deep-learning-the-hardest-go-problem-in-the-world/> https://blog.janestreet.com/deep-learning-the-hardest-go-pro.... This covers developments since the initial post, plus the headline problem (which is still open).