On Tue, 2008-11-18 at 12:02 -0500, Michael Williams wrote: > Weston Markham wrote: > > On Mon, Nov 17, 2008 at 11:13 PM, Michael Williams > > <[EMAIL PROTECTED]> wrote: > >> No one ever alleged that pure AMAF or pure MC was infinitely scalable. > > > > My point is that in many cases, they doesn't even keep all of their > > benefits, after some number of trials have been run. So, running 10k > > playouts can be significantly worse than running 5k playouts. > > It doesn't make any sense to me from a theoretical perspective. Do you have > empirical evidence?
I have never seen any evidence that it gets weaker with the number of playouts. In most normal situations I would think that it's inclined to prefer one specific move and that move is increasingly likely to get selected as you do more playouts. And since this style of bot is relatively weak, that move will not be best. But that is not the same as saying that it will play better in general if you limit the number of playouts. I suspect the strength level approaches some asymptote, in other words it will play better and better with the number of playouts, but will never quite reach some very modest level of play. At some point you may need to add a billion playouts to add 1 ELO point. - Don > > _______________________________________________ > computer-go mailing list > [email protected] > http://www.computer-go.org/mailman/listinfo/computer-go/
signature.asc
Description: This is a digitally signed message part
_______________________________________________ computer-go mailing list [email protected] http://www.computer-go.org/mailman/listinfo/computer-go/
