On Tue, Feb 17, 2009 at 4:39 PM, <dave.de...@planet.nl> wrote: > I've been looking into CGT lately and I stumbled on some articles about > approximating strategies for determining the sum of subgames (Thermostrat, > MixedStrat, HotStrat etc.) > It is not clear to me why approximating strategies are needed. What is the > problem? Is Ko the problem? Is an exact computation too slow? > Can anyone shed some light on this?
I'm sure someone else could give a better answer, but it does come down to slowness. Same thing as assuming the value of a gote move equals the position value if black moves first averaged with its value if white moves first, and the game score equals original score plus half the value of the largest move on the board -- these assumptions are wrong, and they estimates are not guaranteed to yield either the correct score or the biggest play on the board, but you have to do something if you can't perfectly read out the rest of the game. If CGT values were all nice real numbers that summed normally when combining subgames, then CGT would be a lot simpler. The possibility of wanting to play a ko threat in another part of the board prevents the two areas from being separate subgames in the technical sense -- the information sets are shared. _______________________________________________ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
