Le mardi 22 janvier 2008, Petri Pitkanen a écrit : > 2008/1/22, Eric Boesch <[EMAIL PROTECTED]>: > > On Jan 22, 2008 1:43 AM, Petri Pitkanen <[EMAIL PROTECTED]> wrote: > > > Even top MC programs fail to see that a group with 3 liberties with > > > no eyes is dead. > > > > A 3-liberty group with no eyes has a 100% chance to die during > > playouts unless a surrounding group dies first. 100% chance to die is > > as good a job of "seeing" deadness as a generic MC playout can ever > > do. > > > > But As I played these programs they have a completely lost game. A > 3-lib group against a group with so many liberties that i did not > bother to count and it did not resign. Until I got tired of it and > removed a liberty. After that it managed to simulate yje game as > hopelessly lost and resigned . This was crazy stones if memeory > serves. Similar thing have happened with other programs. > > It also leads them to losing semeais because take may take a move in > other part of board when they cannot afford it. Assuming that group > entangled in fight have 3 liberties or more. Which is one trick to win > them "unfairly" > > This is not a bug, its a feature of MC program. It seems you really know very little about their logic and strenght. Try on 9X9 and you will see they are very strong at tactics.
They play to maximize winning probability, not score. So if an MC program wins, it will probably do it by 0.5 point, but if it loses it will try "looking stupid" things, but its the better way to win. On 19x19 they play cosmic style, and are strong at sacrifice. add some handicap if you are stronger , they are good oponent. Alain _______________________________________________ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
