Monthly
288 pp. per issue
6 x 9, illustrated
ISSN
0899-7667
E-ISSN
1530-888X
2014 Impact factor:
2.21

Neural Computation

September 2007, Vol. 19, No. 9, Pages 2492-2514
(doi: 10.1162/neco.2007.19.9.2492)
© 2007 Massachusetts Institute of Technology
Asymptotic Behavior and Synchronizability Characteristics of a Class of Recurrent Neural Networks
Article PDF (168.1 KB)
Abstract

We propose an approach to the analysis of the influence of the topology of a neural network on its synchronizability in the sense of equal output activity rates given by a particular neural network model. The model we introduce is a variation of the Zhang model. We investigate the time-asymptotic behavior of the corresponding dynamical system (in particular, the conditions for the existence of an invariant compact asymptotic set) and apply the results of the synchronizability analysis on a class of random scale free networks and to the classical random networks with Poisson connectivity distribution.