Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: A New Self-Play Experiment -- Diminishing Returns Shown with 95% Conf.

Author: Ernst A. Heinz

Date: 08:28:10 05/26/00

Go up one level in this thread


>I think that Tom means to the following fact:
>depth increases, the number of draws increases.
>
>This can be right also for equal depth.

Sure, it can -- but this is a totally different question than
the one my experiment seeks to answer.

I merely state that the diminishing returns in "X+1 <=> X"
self-play of "Fritz 6" _manifest_ themselves in less wins and
more draws while losses remaining roughly steady.

So, why should I have to do all the "X <=> X" matches then
in the setting of _my_ current experiment?

This is what Tom repeatedly insisted on and I denied.

=Ernst=



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.