Newsgroups: comp.ai.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!oitnews.harvard.edu!purdue!haven.umd.edu!news.umbc.edu!eff!news.duke.edu!news.mathworks.com!tank.news.pipex.net!pipex!demon!sunsite.doc.ic.ac.uk!warwick!bsmail!usenet
From: Max Ratcliffe <m.j.ratcliffe@bris.ac.uk>
Subject: large number of outputs
Message-ID: <DHzKC3.IA3@uns.bris.ac.uk>
Sender: usenet@uns.bris.ac.uk (Usenet news owner)
Nntp-Posting-Host: pc19.aer.bris.ac.uk
Organization: University of Bristol, England
Date: Mon, 13 Nov 1995 14:31:15 GMT
Lines: 14

Hi, 

Does anybody know of any reason why a network with more outputs than 
inputs cannot be used.  I heard somewhere that such a configuration 
was (very loosely) analogous to a system with more unknowns than 
equations, but I cannot really see the analogy.  

I imagine that such a system might be very susceptible to noise, and
might also produce pretty unpredictable results, and I'd be interested
to know if anybody has any thoughts.

Thanks, 

Max
