Thanks for sending the statistics. I'll try them out later on my programs too.
There is only 1 way to resolve this - maybe we should test it out on a 19x19 handicap server. We can play a few weeks and then take a look at the statistics later. I predict that gnugo will perform better on handicap games relative to non-handicap games and monte carlo players. I agree that monte carlo program will do very well when they are given the advantage. My fear is that the weak monte carlo program will do better than they should against the stronger ones when give the advantage of handicap games. I will also predict that your 9 stone 0.74 score will very quickly degrade to close to 1.0 after a few moves. 9 stones is a dead won game and the score is far from accurate at 0.74. Can you try playing a couple of reasonable moves against crazy stone at this handicap and see what happens to the score? - Don On Fri, 2006-12-22 at 17:25 +0100, Rémi Coulom wrote: > Don Dailey wrote: > > Hi Steve, > > > > What you fail to take into considerations is that a monte/carlo > > player may ruin it's chances before the weaker player has a > > chance to play a bad move. The monte carlo player sees all > > moves as losing and will play almost randomly. > I don't agree. Here is the winning percentage I get with Crazy Stone at > various handicaps, with a komi of 0.5, over 10000 random simulations: > > 9 Stones: 0.74 > 8 Stones: 0.73 > 7 Stones: 0.69 > 6 Stones: 0.67 > 5 Stones: 0.63 > 4 Stones: 0.61 > 3 Stones: 0.57 > 2 Stones: 0.54 > > My program still plays reasonable moves at these winning rates. > > I tend to believe MC programs would handle handicap better than pure > territory-based programs, because they know how to play safe when they > are ahead, and risky when behind. > > If my program is much stronger than its opponent, then it will not play > blunders that the opponent can easily take advantage of, whatever the > handicap. > > This being said, I don't believe my program can give handicap to any > other on 19x19 ;-) > > Rémi _______________________________________________ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
