Newsgroups: comp.ai.genetic
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!hookup!swrinde!pipex!uknet!festival!ainews!roberts
From: roberts@aisb.ed.ac.uk (Robert Saunders)
Subject: Re: Definition of Genetic Programming
Message-ID: <Cz9IAt.Gwo@aisb.ed.ac.uk>
Sender: news@aisb.ed.ac.uk (Network News Administrator)
Organization: Dept AI, Edinburgh University, Scotland
X-Newsreader: TIN [version 1.2 PL2]
References: <mreddy-141194082133@mreddy.comp.glam.ac.uk>
Date: Mon, 14 Nov 1994 14:44:52 GMT
Lines: 20

This is only a quick guess but...

Koza cites the use of Genetic Algorithms to create Finite State
Automotas in his book considering this to be an example of Genetic
Algorithms and NOT Genetic Programming. The main difference seems
to be that the representation has to be carefully coded to be of use
to solve the problem. The representation used is still of fixed length.
I don't know much about core wars but I would guess the problem
will need a careful encoding of the problem to be able to use Genetic
Algorithms which is domain specific. Would this be fixed length?
I think that Genetic Programming is not defined by the tree structure
but the generality of the operations upon it without a need for
careful encoding of the problem.

Again it's only a guess and I'm a bit of a newbie

Hope this made some sense

Rob

