On Thu, Feb 11, 2010 at 01:06:34PM +0100, Petr Baudis wrote: > extra_komi = 7.5 * handicap_stones_count > Then it is linearly decreased until it hits 0 at move 200.
All this discussion made me think - has anyone tried to adjust the komi between the simulations. Run (say) 10% of the simulations you expect to run for a move, and see how many of the moves ended in a win. If there are many moves, adjust the komi to make winning harder. If there no moves at all, adjust the komi to make winning easier. Repeat after a small number of simulations (maybe after every one). Yes, this will add noise to the results, when a reasonably good move gets a loss marked against it when the komi gets too hard, but then again, the whole system is noisy already, with random simulations at the leaves of the tree... This sounds like such a simple and obvious idea that someone must have tried it already. If I had a working implementation to play with, I might try it myself, but I don't, and I don't have the time to set one up and try. - Heikki -- Heikki Levanto "In Murphy We Turst" heikki (at) lsd (dot) dk _______________________________________________ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
