Newsgroups: comp.ai.genetic
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!news.mathworks.com!uunet!hearst.acc.Virginia.EDU!murdoch!fulton.seas.Virginia.EDU!jpi6g
From: jpi6g@fulton.seas.Virginia.EDU (James P. Ignizio)
Subject: GAs and Neural Nets
X-Nntp-Posting-Host: fulton.seas.virginia.edu
Message-ID: <D6KMq7.4yJ@murdoch.acc.Virginia.EDU>
Sender: usenet@murdoch.acc.Virginia.EDU
Organization: University of Virginia
Date: Wed, 5 Apr 1995 16:49:19 GMT
Lines: 29

5 April 1995

RE: Using GAs to Optimize Neural Networks and
Misconceptions with Regard to Linear Methods

I have received a large number of requests with regard to
copies of the papers that I mentioned that cover our work
in using linear programming to simultaneously design and
train neural networks (and our more recent work in using
genetic algorithms to accomplish the same thing). Due to the
number of requests, I will be making these papers available
on our department's home page --- and this should be done
by the end of next week. The address is:

http://freney.sys.virginia.edu/homepage.html

I would also like to attempt to clear up an evident
misunderstanding as brought up by Jim Golden of 
Vanderbilt. The use of linear programming does not require
one to *linearize* nonlinear functions. Thus no information
is lost nor is the model compromised.

Jim Ignizio

--
Dr. James P. Ignizio          ignizio@virginia.edu
Professor and Chair           804-924-5394 (office)
Systems Engineering           804-982-2972 (fax)
University of Virginia        Charlottesville, VA 22903
