Newsgroups: comp.ai.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!nntp.club.cc.cmu.edu!hudson.lm.com!news.pop.psu.edu!news.cac.psu.edu!howland.reston.ans.net!cs.utexas.edu!utnut!nott!cunews!avs
From: avs@doe.carleton.ca (Andy Veluswami)
Subject: Don't care states in NN
Message-ID: <Cyt6AG.Av5@cunews.carleton.ca>
Sender: news@cunews.carleton.ca (News Administrator)
Organization: Carleton University
X-Newsreader: TIN [version 1.2 PL2]
Date: Sat, 5 Nov 1994 19:03:52 GMT
Lines: 18

Hi all,

I am trying to train a single neural network to model
two different functions (which are closely related).
Both the functions have the same set of input variables, 
but the second function needs a few extra inputs.  I have
one input variable defining which function to give
at the output ( a switch so to say).   The extra variables are not
used by the first function,  so I was wondering if there is a 'don't 
care' state or some value I could assign to them when I am using the
network to model the first function.

If anyone has any experience in this kind of thing I would really
appreciate it if you could help me out. 

thanks a 1e6,

andy
