Header

UZH-Logo

Maintenance Infos

Existence and Global Attractivity of Stable Solutions in Neural Networks


Leoni, Patrick; Senesi, Pietro (2004). Existence and Global Attractivity of Stable Solutions in Neural Networks. Working paper series / Institute for Empirical Research in Economics No. 198, University of Zurich.

Abstract

The present paper shows that a sufficient condition for the existence of a stable solution to an autoregressive neural network model is the continuity and boundedness of the activation function of the hidden units in the multi layer perceptron (MLP). In addition, uniqueness of a stable solution is ensured by global lipschitzness and some conditions on the parameters of the system. In this case, the stable value is globally stable and convergence of the learning process occurs at exponential rate.

Abstract

The present paper shows that a sufficient condition for the existence of a stable solution to an autoregressive neural network model is the continuity and boundedness of the activation function of the hidden units in the multi layer perceptron (MLP). In addition, uniqueness of a stable solution is ensured by global lipschitzness and some conditions on the parameters of the system. In this case, the stable value is globally stable and convergence of the learning process occurs at exponential rate.

Statistics

Downloads

423 downloads since deposited on 29 Nov 2011
3 downloads since 12 months
Detailed statistics

Additional indexing

Item Type:Working Paper
Communities & Collections:03 Faculty of Economics > Department of Economics
Working Paper Series > Institute for Empirical Research in Economics (former)
Dewey Decimal Classification:330 Economics
Language:English
Date:August 2004
Deposited On:29 Nov 2011 22:32
Last Modified:19 Feb 2018 20:37
Series Name:Working paper series / Institute for Empirical Research in Economics
ISSN:1424-0459
OA Status:Green
Official URL:http://www.econ.uzh.ch/wp.html

Download

Download PDF  'Existence and Global Attractivity of Stable Solutions in Neural Networks'.
Preview
Filetype: PDF
Size: 272kB