Author: James T. Walker
Date: 06:43:31 09/15/00
Go up one level in this thread
On September 14, 2000 at 23:36:49, Dann Corbit wrote: >On September 14, 2000 at 23:30:42, pavel wrote: > >>here is a post from the author of zarkov in a forum >> >>http://satirist.org/learn-game/methods/ga/chess.html >> > >It looks like he concluded it was not worth the bother. > >Interestingly, from Tim Mann's site you can find the code the Deep Blue folks >used to tune evaluation paramters. The most interesting thing about it is that >it is automatic and reacts to a big file of games by GM players. Hello Dann, I think I read about this many years ago. I think this automatic tuning goes way back to Deep Thought in the early days. Maybe Bob can elaborate. It seems the first time I read about this they would let the program analyze 1000 GM games and tune itself to match the most moves of the GM's or something like that. Is this correct? Jim
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.