ecms_neu_mini.png

Digital Library

of the European Council for Modelling and Simulation

 

Title:

On Stability And Model Order Reduction Of Perturbed

Nonlinear Neural Networks

Authors:

Marissa Condon, Georgi G. Grahovski

Published in:

 

ECMS 2008 Proceedings

Edited by: Loucas S. Louca, Yiorgos Chrysanthou, Zuzana Oplatkova, Khalid Al-Begain

 

ISBN: 978-0-9553018-6-5

Doi: 10.7148/2008

 

22nd European Conference on Modelling and Simulation,

Nicosia, June 3-6, 2008

 

Citation format:

Condon, M., & Grahovski, G. G. (2008). On Stability And Model Order Reduction Of Perturbed Nonlinear Neural Networks. ECMS 2008 Proceedings edited by: L. S. Louca, Y. Chrysanthou, Z. Oplatkova, K. Al-Begain (pp. 292-298). European Council for Modeling and Simulation. doi:10.7148/2008-0292

DOI:

http://dx.doi.org/10.7148/2008-0292

Abstract:

In this paper, the qualitative theory of large-scale dynam- ical systems is surveyed. In particular, the focus is the Hopfield Neural networks both with and without pertur- bations. Properties relating to asymptotic and exponen- tial stability and instability are detailed. A model reduc- tion technique based on balanced truncation is applied to the neural networks. Its effect on the stability properties of the networks is then examined. A numerical test illus- trates some important points.

Full text: