From: Łukasz Lew <[email protected]>
>I was wondering what are the good (fast/accurate) ways of evaluating
program strength.
>The most accurate one is to play many games against gnugo or on KGS.
>But it is quite slow as many games are needed.
>Another one is to have set of labeled positions (win/loss) and make
>your program predict the labels. (This is what MoGo guys did)
>It is much faster. But how well it is correlated with the true strength?
If your program were 100% accurate, it would win all winnable games.
When any program loses a game, it would be interesting to use deeper search to
backtrack, find the blunder which lost the game, and add key positions to the
set. Tune your program so that it knows that move A loses, move B wins.
_______________________________________________
computer-go mailing list
[email protected]
http://www.computer-go.org/mailman/listinfo/computer-go/
