News for hackers. Stuff that matters.
Re:Deep learning for chess by thomasahle on Tuesday December 02, 2014 @08:57AM (#8687052)
Actually it just might be possible to do, at least in a probabilistic sense. The author of Rybka at least managed to 'prove' (in a weak sense) that a certain opening is unplayable. Quite facinating: http://en.chessbase.com/post/rajlich-busting-the-king-s-gamb...
Re:Deep learning for chess by ajkjk on Tuesday December 02, 2014 @09:25AM (#8687201)
That article is an April Fool's joke. See http://en.chessbase.com/post/the-chebase-april-fool-s-prank.
Re:Deep learning for chess by thomasahle on Tuesday December 02, 2014 @10:20AM (#8687502)
A second of April joke. Cheeky. The follow up comments from Vas Rajlich are still interesting though.
Re:Deep learning for chess by slm_HN on Tuesday December 02, 2014 @12:59PM (#8688588)
Vas Rajlich could write a program that solved chess. He would simply have to find an existing program that solved chess and copy it.