Premium
Robustness of convergence in finite time for linear programming neural networks
Author(s) -
Marco Mauro Di,
Forti Mauro,
Grazzini Massimo
Publication year - 2006
Publication title -
international journal of circuit theory and applications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.364
H-Index - 52
eISSN - 1097-007X
pISSN - 0098-9886
DOI - 10.1002/cta.352
Subject(s) - robustness (evolution) , artificial neural network , convergence (economics) , linear programming , computer science , mathematical optimization , class (philosophy) , control theory (sociology) , mathematics , artificial intelligence , biochemistry , chemistry , economics , gene , economic growth , control (management)
A recent work has introduced a class of neural networks for solving linear programming problems, where all trajectories converge toward the global optimal solution in finite time. In this paper, it is shown that global convergence in finite time is robust with respect to tolerances in the electronic implementation, and an estimate of the allowed perturbations preserving convergence is obtained. Copyright © 2006 John Wiley & Sons, Ltd.