Newsgroups: comp.ai.genetic
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!news.alpha.net!uwm.edu!news.moneng.mei.com!howland.reston.ans.net!pipex!sunic!news.chalmers.se!news.gu.se!gd-news!d6241.shv.hb.se!sa209
From: sa209@utb.shv.hb.se (Claes Andersson)
Subject: Re: dynamic changing of fitness function
Message-ID: <sa209.70@utb.shv.hb.se>
Sender: usenet@gdunix.gd.chalmers.se (USENET News System)
Nntp-Posting-Host: d6241.shv.hb.se
Organization: Department of Scocial Science
References: <ECS9411151353A@pine.shu.ac.uk>
Date: Tue, 15 Nov 1994 21:00:03 GMT
Lines: 22

In article <ECS9411151353A@pine.shu.ac.uk> cewright@pine.shu.ac.uk (Chris Wright) writes:

> I'm currently working on my degree project and it involves
> the dynamic alteration of the fitness function (this depends 
> on the user's input into the system.)

> Anyone out there had any experience/thoughts on this?

> Chris Wright

> cewright@pine.shu.ac.uk


	The dynamic fitness function I've used was very simple and simulated 
a biotope with variable dynamics. What happens is that a highly converged 
population is more likely to go distinct at small alterations of the fitness 
function than loosely adapted ones. Another model is to let the fitness 
function be a trait of the animats... then it gets really intracate. I haven'
t used it but I imagine that it will look like games theory since a specimen 
affects its own fitness function and therefore must adapt accordingly.

Claes Andersson. University of Bors. Sweden.
