[Computer-go] I need an off-the-shelf final position live/dead evaluator

Brian Sheppard sheppardco at aol.com
Mon Nov 29 14:30:07 PST 2010


>Yes but once you know which stones you can disregard as being dead,
>counting the board becomes a simple flood-fill exercise, isn't it?

I was thinking the same thing: basically play out the position using UCT. If
you have a few hundred trials UCT, you might get very accurate play. A
downside: you might have to play many moves to reach a terminal position,
and your UCT must prefer high point differential.

Or just use one search of a thousand UCT trials, where we keep track of how
often a point is black/white at the end of the game. Then assume that stones
that live with probability < T are dead.

My guess is that either of these approaches can be made to work to the
standard that Dave Dyer suggests: usually correct to within a few points.

Brian





More information about the Computer-go mailing list