Designing the Hopfield neural network via pole assignment
Authors:
- 1 April 1994
- journal article
- research article
- Published by Taylor & Francis in International Journal of Systems Science
- Vol. 25 (4) , 669-681
Abstract
We consider the stability and design of the general non-symmetric Hopfield neural network. The stability of equilibria is analysed through the linearized system and a stability criterion is derived. Bounds of the equilibria are obtained and can be used to start off iterative processes in the computation of the equilibria. The design problem is shown to be equivalent to an inverse eigenvalue problem and is analysed using pole assignment techniques in control theory. The novel and efficient algorithm for the design of Hopfield neural networks with a given equilibrium is then presented. Two illustrative examples are given.Keywords
This publication has 6 references indexed in Scilit:
- Introduction to the Theory of Neural ComputationAmerican Journal of Physics, 1994
- Qualitative analysis of neural networksIEEE Transactions on Circuits and Systems, 1989
- Convergent activation dynamics in continuous time networksNeural Networks, 1989
- On the stability, storage capacity, and design of nonlinear continuous neural networksIEEE Transactions on Systems, Man, and Cybernetics, 1988
- Neurons with graded response have collective computational properties like those of two-state neurons.Proceedings of the National Academy of Sciences, 1984
- Absolute stability of global pattern formation and parallel memory storage by competitive neural networksIEEE Transactions on Systems, Man, and Cybernetics, 1983