On Thu, 2008-12-18 at 03:27 +0000, Weston Markham wrote: > On Mon, Dec 15, 2008 at 11:10 PM, Mark Boon <tesujisoftw...@gmail.com> wrote: > > It would have been much more persuasive if you had simply run a 5K > > playout bot against a 100K bot and see which wins more. > > In 200 games, 100k beat 5k a total of 127 times. So that's about a > 63.5% win rate.
That's better than I thought it would be. But I would guess that most of the benefit would be seen in doing just 10 or 20k playouts. This is because after 5K you are really in the seriously diminishing returns part of the curve. With a 63.5% win rate there is still a great deal of uncertainty, but this is evidence that it's probably better - which of course is very believable since that is what we would expect. - Don > _______________________________________________ > computer-go mailing list > computer-go@computer-go.org > http://www.computer-go.org/mailman/listinfo/computer-go/ _______________________________________________ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/