By 2030, AI can defeat the best Go players, training only on as many games as the best Go players have played.
➕
Plus
11
Ṁ271
2030
47%
chance

For reference, DeepMind’s AlphaGo has probably played a hundred million games of self-play, while Lee Sedol has probably played 50,000 games in his life

Get
Ṁ1,000
and
S3.00
Sort by:

Should do one on whether such an AI you elude to, whether there will be something like that that’s open-source. :O

© Manifold Markets, Inc.Terms + Mana-only TermsPrivacyRules