Newsgroups: comp.ai.genetic
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!oitnews.harvard.edu!purdue!lerc.nasa.gov!magnus.acs.ohio-state.edu!math.ohio-state.edu!howland.reston.ans.net!newsfeed.internetmci.com!tank.news.pipex.net!pipex!demon!peer-news.britain.eu.net!lyra.csx.cam.ac.uk!warwick!bsmail!usenet
From: David Brittain <david.brittain@bris.ac.uk>
Subject: Re: Sub-population sand diversity
Content-Type: text/plain; charset=us-ascii
To: Patrick Murphy <103120.75@compuserve.com>
Message-ID: <311F0B1C.1383@bris.ac.uk>
Sender: usenet@uns.bris.ac.uk (Usenet news owner)
Nntp-Posting-Host: dig1.enm.bris.ac.uk
Content-Transfer-Encoding: 7bit
Organization: University of Bristol
References: <4fll4o$q5b@dub-news-svc-4.compuserve.com>
Mime-Version: 1.0
Date: Mon, 12 Feb 1996 09:40:44 GMT
X-Mailer: Mozilla 2.0 (Win16; I)
Lines: 25

Patrick Murphy wrote:
> 
> In real-life evolution species tend to get monopolies on the
> environment.  In other words, once mammals are firmly established
> it is difficult for other types of species to evolve in any of the
> mammal niches.  Isolated environments such as islands allow the
> evolution of new species that don't need to compete with the
> established mainland species. Has anyone tried to breed GA's in
> isolated sub-populations?

There has been work done on this.  A number of populations are set up and 
allowed to independently evolve.  Every so often individuals migrate to 
between the populations.  This technique has proved more computationally 
efficient than GA's using a single population.  There are a number of 
references conatined in :
Thomas Back & Frank Hoffmeister, Basic Aspects of Evolution Strategies, 
Statistics and Computing, 1994, Vol. 4, pp51-63.

Dave.
------------------------------------------------------------------------David Brittain					david.brittain@bris.ac.uk
Design Information Group			(44) 117 928 8913
Dept. of Engineering Mathematics
University of Bristol
BRISTOL
------------------------------------------------------------------------
