Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!uwm.edu!linac!convex!convex!dyes
From: dyes@convex.convex.COM (Tim Dyes)
Newsgroups: comp.ai.neural-nets
Subject: Re: Backprop with additional noisy inputfeature
Message-ID: <1991May14.182932.21193@convex.com>
Date: 14 May 91 18:29:32 GMT
References: <3559@wn1.sci.kun.nl>
Sender: dyes@convex (Tim Dyes)
Organization: CONVEX Computer Corporation
Lines: 17
Nntp-Posting-Host: aspen.convex.com

In article <3559@wn1.sci.kun.nl>, mderksen@sci.kun.nl (M. Derksen) writes:
|> Dear neural netters,
|> 
|> Can someone explain me why an additional noisy inputfeature results in a
|> better prediction performance (generalization). 
|> 

Try this on for size...
Adding a noisy input feature in effect adds a noise term to the input of each
node within the 1st hidden layer.  Successful training forces the network 
to adjust its weights to overcome this noise level.  That will result in there
being more distance between the 1st hidden layer's output vectors that
correspond to different network outputs.  This in effect raises the 
discrimination capability of the 3rd layer to the 2nd layer's output, and so
allows it to more correctly classify an input pattern.

- Tim Dyes