Title :
A stability criterion for Hopfield networks based on Popov theorem
Author_Institution :
Dept. of Electr. Eng., Concepcion Univ.
fDate :
27 Jun-2 Jul 1994
Abstract :
A Hopfield network is considered as a nonlinear system with multiples nonlinearities (so called Lurie systems). A frequency domain criterion is formulated to prove global asymptotic stability of the equilibrium point of the system, this condition is expressed as a inequality over the whole range of frequency. This known result is applied to analyse the global asymptotic stability of a Hopfield network, obtaining conditions for testing the stability of the equilibrium points and their regions of attractions. These conditions do not require symmetric interconnection matrices
Keywords :
Hopfield neural nets; Popov criterion; asymptotic stability; frequency-domain analysis; nonlinear systems; Hopfield neural networks; Lurie systems; Popov theorem; equilibrium point; frequency domain criterion; global asymptotic stability; inequality; nonlinear system; nonlinearities; stability criterion; Asymptotic stability; Equations; Frequency domain analysis; Gold; Linear matrix inequalities; Nonlinear systems; Stability analysis; Stability criteria; Symmetric matrices; Testing;
Conference_Titel :
Neural Networks, 1994. IEEE World Congress on Computational Intelligence., 1994 IEEE International Conference on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-1901-X
DOI :
10.1109/ICNN.1994.375010
