DocumentCode
1333338
Title
Global asymptotic stability of a class of dynamical neural networks
Author
Arik, Sabri
Author_Institution
Dept. of Electron., Istanbul Univ., Turkey
Volume
47
Issue
4
fYear
2000
fDate
4/1/2000 12:00:00 AM
Firstpage
568
Lastpage
571
Abstract
In this paper, we present a sufficient condition for the existence, uniqueness, and global asymptotic stability (GAS) of the equilibrium point for a class of dynamical neural networks. It is shown that the quasi-diagonally column-sum dominant condition on the interconnection matrix of the neural network proves the existence, uniqueness, and GAS of the equilibrium point with respect to all nondecreasing activation functions. This condition is also compared with the previous results derived in the literature
Keywords
asymptotic stability; neural nets; transfer functions; dynamical neural networks; equilibrium point; global asymptotic stability; interconnection matrix; nondecreasing activation functions; quasi-diagonally column-sum dominant condition; Asymptotic stability; Circuits; Design optimization; Equations; Neural networks; Stability analysis; Sufficient conditions;
fLanguage
English
Journal_Title
Circuits and Systems I: Fundamental Theory and Applications, IEEE Transactions on
Publisher
ieee
ISSN
1057-7122
Type
jour
DOI
10.1109/81.841858
Filename
841858
Link To Document