Abstract
A stability criterion for the Hopfield neural networks with continuous and discontinuous nonlinearities and operation described by nonlinear differential equations is formulated and is based on the sign of the logarithmic norm of a matrix.
Similar content being viewed by others
REFERENCES
Gorban', A.N. and Rossiev, D.A., Neironnye seti na personal'nom komp'yutere (Neural Networks for Personal Computers), Novosibirsk: Nauka, 1996.
Daletskii, Yu.L. and Krein, M. G., Ustoichivost' reshenii differentsial'nykh uravnenii v banakhovom prostranstve (Stability of the Solutions of Differential Equations in the Banach Space), Moscow: Nauka, 1970.
Filippov, A.F., Differentsial'nye uravneniya s razryvnoi pravoi chast'yu, Moscow: Nauka, 1985. Translated under the title Differential Equations with Discontinuous Right Side, Dordrecht: Kluwer, 1988.
Hopfield, J.J., Neurons with Graded Response Have Collective Computational Properties Like Those of Two-State Neurons, 1984, vol. 81, pp. 3088-3092.
Dudnikov, E.E. and Rybashov, M.V., Absolute Stability of a Class of Feedback Neural Networks, Avtom. Telemekh., 1999, no. 12, pp. 33-40.
Borisov, Yu.I., Kuz'michev, Yu.A., Smolitskii, Kh.P., et al., Equilibrium States in Continuous Hopfield Neural Network Models, Priborostroenie, 1994, no. 3-4, pp. 5-12.
Petryakova, E.A., Hopfield Neural Networks with Skew Neuron-Neuron Coupling Matrix, Priborostroenie, 1994, vol. 37, no.3-4, pp. 24-32.
Demidovich, B.L., Lektsii po matematicheskoi teorii ustoichivosti (Lectures on Mathematical Theory of Stability), Moscow: Nauka, 1967.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Boikov, I.V. Stability of Hopfield Neural Networks. Automation and Remote Control 64, 1474–1487 (2003). https://doi.org/10.1023/A:1026056104067
Issue Date:
DOI: https://doi.org/10.1023/A:1026056104067