Re: [computer-go] Published source for mercy rule?

2009-03-01 Thread Łukasz Lew
mercy rule in libego: http://github.com/lukaszlew/libego/blob/77b4dd4e035e4b44c17204557c9930a52e10e0c0/ego/playout.h line 55. Regards, Łukasz Lew On Fri, Feb 27, 2009 at 01:00, Seth Pellegrino wrote: > Hello list, > > I've managed to track the idea of a mercy rule in monte-carlo playouts back >

Re: [computer-go] Published source for mercy rule?

2009-03-01 Thread Don Dailey
Has anyone done a good analysis on the value of the mercy rule? It is a major or minor speedup? And how does it affect the quality of the search? Is it more useful with bigger board sizes? It seems like I remember seeing that many believed it to be a very minor improvement if any, but I'm

Re: [computer-go] Published source for mercy rule?

2009-03-01 Thread dhillismail
Bruno Bouzy gives it credit for a 1% strength improvement in the tutorial: "Old-fashioned Computer Go vs Monte-Carlo Go" http://www.math-info.univ-paris5.fr/~bouzy/publications/CIG07-tutorial-1avril2007.pdf (It's a pretty good read in general, btw.) The Mercy rule interacts with ownership maps and

RE: [computer-go] re-CGT approximating the sum of subgames

2009-03-01 Thread Philipp Hennig
Hi Martin, Dave and others, I have never posted to this list before (so, hi everyone! :-), but your discussion on approximate CGT caught my eye. Dave. you mentioned that you are interested in trying to mix UCT and CGT: Dave Devos wrote on 19 Feb 2009, 19:33 GMT: Ok, I'll look into your tem