Newsgroups: comp.ai.genetic
From: ward@mondas.demon.co.uk (Peter Ward)
Path: cantaloupe.srv.cs.cmu.edu!das-news.harvard.edu!news2.near.net!MathWorks.Com!europa.eng.gtefsd.com!howland.reston.ans.net!swrinde!pipex!demon!mondas.demon.co.uk!ward
Subject: Re: "a wondrous(?) computation"
References: <dehaan-1209941313500001@dehaan1.psych.kun.nl> <3549sj$1lqu@tornews.torolab.ibm.com>
Organization: Mondas
Reply-To: ward@mondas.demon.co.uk
X-Newsreader: Demon Internet Simple News v1.27
Lines: 24
Date: Wed, 14 Sep 1994 11:14:33 +0000
Message-ID: <779541273snz@mondas.demon.co.uk>
Sender: usenet@demon.co.uk

In article <3549sj$1lqu@tornews.torolab.ibm.com>
           marcap@marcap.torolab.ibm.com "Marc Pawlowsky" writes:

> >A fuzzy problem. A student of mine described a program he encountered. It
> >is said to be able to discover functions from a set of IO data. He also
> >told it was some kind of GA. Can someone give me a hint that I can pass
> >over to him to find such a tool, or to construct it from a know algorithm?
> 
> If you student has an idea of what the function should look like
> then a GA could be used to find the function parameters.

Genetic Programming (GP) may be able to derive the function
even if you have _no_idea_ what it looks like. Try and look at
John Koza's book - Genetic Programming (pub: MIT press,
ISBN: 0-262-11170-5. The University of Texas have an ftp site
(cc.utexas.edu ???) carrying some source code in various
languages that do exactly what your student describes.

-- 

Pete Ward                   I know it's irrational but at times
Mondas IT Ltd               of stress I take great comfort from
                            the belief that somewhere out there
                            there really is a free lunch.
