Newsgroups: comp.ai.neural-nets,swnet.ai.neural-nets,triangle.neural-nets,uw.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!oitnews.harvard.edu!news.cs.umb.edu!CTCnet!news.math.psu.edu!feeder.chicago.cic.net!news.indiana.edu!vixen.cso.uiuc.edu!news.sprintlink.net!news-peer.sprintlink.net!interpath!news.interpath.net!news.interpath.net!sas!newshost.unx.sas.com!saswss
From: saswss@hotellng.unx.sas.com (Warren Sarle)
Subject: Re: Treshold question: please help
Originator: saswss@hotellng.unx.sas.com
Sender: news@unx.sas.com (Noter of Newsworthy Events)
Message-ID: <E4F4zC.A64@unx.sas.com>
Date: Wed, 22 Jan 1997 16:40:23 GMT
X-Nntp-Posting-Host: hotellng.unx.sas.com
References:  <5bvle6$f2s@news.istar.ca>
Organization: SAS Institute Inc.
Followup-To: comp.ai.neural-nets
Lines: 21


In article <5bvle6$f2s@news.istar.ca>, stepanv@istar.com writes:
|> Please help. If possible, please duplicate an answer via E.mail to:
|> stepanv@istar.ca.

|> The question applies to Feedforward Backpropagation Multilayer NN. 
|> What do we need (and do we need ?) the treshold for? As I understand, the
|> treshold could be represented as an extra input with fixed signal -1 and weight
|> w. 

The usual term is "bias". See "Why use a bias input?" in the
Neural Network FAQ, part 2 of 7: Learning, at
ftp://ftp.sas.com/pub/neural/FAQ2.html

-- 

Warren S. Sarle       SAS Institute Inc.   The opinions expressed here
saswss@unx.sas.com    SAS Campus Drive     are mine and not necessarily
(919) 677-8000        Cary, NC 27513, USA  those of SAS Institute.
 *** Do not send me unsolicited commercial or political email! ***

