Title :
On absolute stability of delayed neural networks
Author :
Sun, Guigen ; Sun, Changyin
Author_Institution :
Res. Inst. of Autom., Southeast Univ., Nanjing, China
fDate :
29 June-1 July 2002
Abstract :
This paper first investigates the absolute exponential stability (AEST) of delayed neural networks with a general class of partially Lipschitz continuous and monotone increasing activation functions. The main results obtained are that if the interconnection matrix T of the delayed neural networks satisfies that -T is an H -matrix with nonnegative diagonal elements and there exists k satisfying the condition concerned, then the neural network system is absolutely stable (ABST).
Keywords :
absolute stability; asymptotic stability; neural nets; transfer functions; absolute exponential stability; delayed neural networks; interconnection matrix; monotone increasing activation functions; neural network system; nonnegative diagonal elements; partially Lipschitz continuous activation functions; Asymptotic stability; Automation; Delay effects; Educational institutions; Erbium; Integrated circuit interconnections; Neural networks; Neurons; Quadratic programming; Sun;
Conference_Titel :
Communications, Circuits and Systems and West Sino Expositions, IEEE 2002 International Conference on
Print_ISBN :
0-7803-7547-5
DOI :
10.1109/ICCCAS.2002.1179099
