Title :
Two alternative stability criteria for discrete-time RMLP
Author :
Barabanov, Nikita E. ; Prokhorov, Dahil V.
Author_Institution :
Dept. of Software Eng., St. Petersburg Electrotechnical Univ., Russia
Abstract :
Two criteria for analysis of global asymptotic stability of recurrent multilayered perceptrons (RMLP) are presented in the unperturbed setting. Both criteria are in the form of LMI. They differ by their strength and computational complexity.
Keywords :
absolute stability; asymptotic stability; discrete time systems; linear matrix inequalities; multilayer perceptrons; recurrent neural nets; stability criteria; state-space methods; LMI; absolute stability; direct method; discrete-time recurrent multilayered perceptrons; global asymptotic stability; linear matrix inequalities; stability criteria; state space extension method; Asymptotic stability; Computational complexity; Equations; Fuzzy logic; Laboratories; Linear matrix inequalities; Multilayer perceptrons; Recurrent neural networks; Stability criteria; Vectors;
Conference_Titel :
Decision and Control, 2002, Proceedings of the 41st IEEE Conference on
Print_ISBN :
0-7803-7516-5
DOI :
10.1109/CDC.2002.1184780
