Author: Rafael Andrist
Date: 06:05:42 11/21/01
Go up one level in this thread
On November 20, 2001 at 14:47:10, Sven Reichard wrote: >Although there is much literature out there on tree search and related things, I >haven't been able to find anything on the design of evaluation functions (except >for the crafty sources :). Most papers on parameter tuning and/or learning >assume a given set of features -- and they deal with Othello. > >Can anybody give me some pointers, or is there just nothing out there? (I can >always dig into the works of Nimzowich, Kotov, and Silman). > >Thanks in advance, >Sven. I didn't found any literature about that subject. For my evaluation function, I studied several books about positional play, pawn structures and endgames written for humans. It's also more interesting to try to implement all these ideas without having a clear how-to-do-list. A brief overview of a possible evaluation function you can find in Shannons paper (http://www.pi.infn.it/~carosi/shannon.txt). Rafael B. Andrist
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.