Science-advisor
REGISTER info/FAQ
Login
username
password
     
forgot password?
register here
 
Research articles
  search articles
  reviews guidelines
  reviews
  articles index
My Pages
my alerts
  my messages
  my reviews
  my favorites
 
 
Stat
Members: 2598
Articles: 1'959'068
Articles rated: 2571

30 May 2020
 
  » » arxiv » 169836

  Article forum


On model selection and the disability of neural networks to decompose tasks
Marc Toussaint ;
Date 19 Feb 2002
Journal Proceedings of the International Joint Conference on Neural Networks (IJCNN 2002), 245-250.
Subject Adaptation and Self-Organizing Systems; Neural and Evolutionary Computing; Disordered Systems and Neural Networks | nlin.AO cond-mat.dis-nn cs.NE
AbstractA neural network with fixed topology can be regarded as a parametrization of functions, which decides on the correlations between functional variations when parameters are adapted. We propose an analysis, based on a differential geometry point of view, that allows to calculate these correlations. In practise, this describes how one response is unlearned while another is trained. Concerning conventional feed-forward neural networks we find that they generically introduce strong correlations, are predisposed to forgetting, and inappropriate for task decomposition. Perspectives to solve these problems are discussed.
Source arXiv, nlin.AO/0202038
Services Forum | Review | PDF | Favorites   
 

No message found in this article forum.  You have a question or message about this article? Ask the community and write a message in the forum.
If you want to rate this article, please use the review section..

Subject of your forum message:
Write your forum message below (min 50, max 2000 characters)

2000 characters left.
Please, read carefully your message since you cannot modify it after submitting.

  To add a message in the forum, you need to login or register first. (free): registration page






ScienXe.org
» my Online CV
» Free


News, job offers and information for researchers and scientists:
home  |  contact  |  terms of use  |  sitemap
Copyright © 2005-2020 - Scimetrica