Author: Pat King
Date: 23:19:07 08/03/98
Go up one level in this thread
On August 03, 1998 at 16:57:04, James Long wrote: >As I rewrite my program, I've been seriously considering >trying a genetic algorithm to train the evaluation >function. I've got a pretty good grasp on how to >handle this, but a couple of things are still >bothering me: > >Most programmers hand tune their eval functions, either >themselves or with the help of a strong player. Is >this because the results are generally better, or >because a good learning eval is difficult to write? The learning part isn't difficult. I had a GA tuning v2.0 of my program several years ago. The problem is that the learning is SLOW! This was compounded in my case by a pretty doggy chess engine. Now that I've V4.0 almost running, I may give it another try. >For the program to "learn," the eval variables must >mutate dependent on the outcome of a game. One approach >I read about is to set up a tournament of 100 "players," >each player having random values assigned to its >eval terms (with the exception of the pawn score). I strongly suggest getting in the ballpark first with some manual tuning, even if you're a lousy chessplayer. Good Luck! Pat
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.