Newsgroups: comp.ai.alife
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!howland.reston.ans.net!EU.net!uknet!festival!leeds.ac.uk!news
From: cgzacb@scs.leeds.ac.uk (A C Black)
Subject: Re: Evolution Maker
Message-ID: <1994Nov2.104653.26949@leeds.ac.uk>
Sender: news@leeds.ac.uk
NNTP-Posting-Host: csgi40.leeds.ac.uk
Organization: The University of Leeds, School of Computer Studies
Date: Wed, 2 Nov 1994 10:46:53 +0000 (GMT)
References: <Cy8M95.8t9@udcf.glasgow.ac.uk>
Lines: 43

In article <Cy8M95.8t9@udcf.glasgow.ac.uk>, richard@milkyway.gla.ac.uk (Richard Lyall) writes:
|> Dear All ...
|> 
|> I have begun writing a piece of software that mimics evolution. It
|> involves creating 'individuals' and giving them 'genes' that determine
|> their phenotype. They then move randomly around and encounter each
|> other, whereupon they mate and produce 'babies' whose genes are derived
|> from their own. At the moment, the individuals cannot learn, but this
|> is coming. They will interact with each other and the environment and
|> gain or lose from such interactions.
|> 
|> Anyone else doing something similar ?
|> 
|> Regards,
|> Richard
|> 
|> ---
|> 

For my final year project I am writing something very similar.  The bugs wander
around in an environment where food appears randomly.  They have genes to
determine movement patterns, range and quality of senses, and speed.  The
program will soon allow bugs to evolve into predators, and I also intend to
include a 'body' gene that determines how the bug appears to other bugs.  It is
forced to change as the other genes change (e.g. a faster bug (change in speed
gene) would have differently shaped 'legs' (represented by a change in the body
gene)) but it will also be able to mutate on it's own - to allow mimicking.  
In writing this I am interested to see if any simple social behaviours emerge.

I would be interested to hear, from any one who has written anything like these
'evolution makers', about the speed with which their bugs have evolved.  My
main worry about writing such a program is that any emergent behavours will
take a long time to appear in the population, so I would like to hear others
experiences in this area.  Previously I have used asexual reproduction, relying
purley on mutation to produce new organisms but perhaps using sexual
reproduction might speed things up?

Andy.
-----
Cognitive Science.
University of Leeds, England.
cgzacb@scs.leeds.ac.uk
csp2acb@lucs-03.novell.leeds.ac.uk
