News for hackers. Stuff that matters.

w0bb13

Re:AlphaGo beats Lee Sedol again in match 2 of 5 on Thursday March 10, 2016 @09:54AM (#11259494)
Alphago isn't programmed to think of that as a mistake (i.e. the team wouldn't be disappointed if she did it). It really only cares about W/L.
Re:AlphaGo beats Lee Sedol again in match 2 of 5 on Thursday March 10, 2016 @09:51AM (#11259475)
I can't believe they would be told not to, or that Lee would abide by it in this match!

Now, I'm not good enough to evaluate if this seems likely, but I could understand that AlphaGo (especially if it was winning) avoided Kos, since they are high variance situations.

Re:Google achieves AI 'breakthrough' by beating Go champion on Tuesday February 02, 2016 @12:31PM (#11020876)
Looks like the other way round - single machine alphago wins 77% of the time, when it gives 4 stones handicap.
Re:A Chess Firewall at Zero? on Tuesday February 02, 2016 @09:03AM (#11019439)
The number of serious games above 19x19 is basically nil, so the difference is a bit misleading (although it's definitely more common in go).
Re:Show HN: Play Go Against a Deep Neural Network on Friday November 13, 2015 @09:14AM (#10559921)
Ditto to me. Seems strange - maybe white tried suicide?
Re:Show HN: Play Go Against a Deep Neural Network on Friday November 13, 2015 @06:55AM (#10559318)
But it doesn't seem possible on the web link for the player to pass and then see the score?