Author: Robert Hyatt
Date: 17:32:43 04/17/00
Go up one level in this thread
On April 17, 2000 at 15:45:06, Mogens Larsen wrote: >The reason I ask is, that in a testmatch against Amy (no learning), Crafty won a >match and when proceded to make another choice in the same opening. The result >was a loss. I have quite a few example of that. Is there a scope for the >learning function? Ie. the first 30 moves or so. > >Best wishes... >Mogens If you _really_ want to see learning affect the games, use the bookw learn 1.0, and set all other weights to .1... then it will play the same game over and over if it wins. I don't set the default values to that as I doubt anyone really wants to play like that...
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.