[Computer-go] Mastering the Game of Go with Deep Neural Networks and Tree Search

Hideki Kato hideki_katoh at ybb.ne.jp
Mon Feb 1 04:44:51 PST 2016


Olivier Teytaud: <CAMpyiGPsNRFtNtk_m1jTvDm4s-_XLcyKGV=SxVC8TiKG9f=Fqg at mail.gmail.com>: 
>Ok, it's not blitz according to http://senseis.xmp.net/?BlitzGames
>(limit at 10s/move for Blitz). But really shorter time settings.
>
>I've seen (as you all) many posts guessing that AlphaGo will lose, but I
>find
>that hard to know. If Fan Hui had won one game, I would say that AlphaGo is
>not ready for Lee Sedol, but with 5-0...

Main time is not so important if we evaluate AI's playing 
strength because AI's thinking time is almost 30 seconds in 
either time setting, due to not-so-smart time control algorithms.
#If the team has developed smarter one, this is wrong.

Anyway, I wonder why Google don't publish the records of the 
informal games.  Certainly losing games are much more 
informative.

Hideki

>(incidentally, there is one great piece of news for machine learning
>people: people in industry are much more interested than before for letting
>us try our deep learning algorithms on their data and that's good for the
>world :-) )
>---- inline file
>_______________________________________________
>Computer-go mailing list
>Computer-go at computer-go.org
>http://computer-go.org/mailman/listinfo/computer-go
-- 
Hideki Kato <mailto:hideki_katoh at ybb.ne.jp>



More information about the Computer-go mailing list