Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Zappa Report

Author: Ingo Althofer

Date: 06:12:37 01/01/06

Go up one level in this thread


On December 31, 2005 at 13:40:15, Vincent Diepeveen wrote:
>On December 30, 2005 at 13:02:43, Ingo Althofer wrote:
>
>Ingo, how do you plan to *ever* win a tournament with a limited amount of
>parameters?

It might be possible, see Fruit as a candidate program.

Also a good example is "Zillions of Games". In many
chess variants this universal program with its
straightforward evaluation of mateirl and mobility
plays surprisingly strong.

There is also a way out for those who love to swim in
zillions of parameters: The theorem deals only with
linear evaluation functions. So, it may not hold in all
cases of nonlinear evals - although common sense
suggests that also there Occams razor should rule.


>>...
>>("On telescoping linear evaluation functions") in the
>>ICCA Journal (now ICGA Jornal), Vol 16 (June 1993),
>>pp. 91-94, describing a theorem (of existence) which says
>>that in case of linear evaluation functions with lots
>>of terms there is always a small subset of the terms
>>such that this set with the right parameters is
>>almost as good as the full evaluation function.

By the way, I never wrote a chess program.
Some Ph.D. students are ways better in doing this.

Ingo.



This page took 0.04 seconds to execute

Last modified: Thu, 07 Jul 11 08:48:38 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.