Author: Christophe Theron
Date: 14:52:00 10/21/99
Go up one level in this thread
On October 21, 1999 at 13:19:45, blass uri wrote:
>On October 21, 1999 at 12:23:58, Christophe Theron wrote:
>
><snipped>
>>* either use their programs for analysis, and also in this case learning is
>>useless.
>
>In this case learning to avoid playing the same opening again is useless but
>learning from previous search when you go forward and backward is not useless.
>
>Uri
You are right. That's why I'll work on "position" learning rather than on "book"
learning.
This way the program will "learn" better. It might take a long time to improve,
but in the end it will be more and more difficult to beat the program, both when
it is in book, and when it is not in book anymore.
Christophe
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.