title: Why feed-forward networks are in a bad shape
creator: Smagt, P. van der
creator: Hirzinger, G.
subject: Neural Nets
subject: Statistical Models
description: It has often been noted that the learning problem in feed-forward neural networks is very badly conditioned. Although, generally, the special form of the transfer function is usually taken to be the cause of this condition, we show that it is caused by the manner in which neurons are connected. By analyzing the expected values of the Hessian in a feed-forward network it is shown that, even in a network where all the learning samples are well chosen and the transfer function is not in its saturated state, the system has a non-optimal condition. We subsequently propose a change in the feed-forward network structure which alleviates this problem. We finally demonstrate the positive influence of this approach.
publisher: Springer Verlag, Berlin
date: 1998
type: Conference Paper
type: NonPeerReviewed
format: application/postscript
identifier: http://cogprints.org/495/2/SmaHir98b.ps
identifier:   Smagt, P. van der and Hirzinger, G.  (1998) Why feed-forward networks are in a bad shape.  [Conference Paper]    (In Press)  
relation: http://cogprints.org/495/