Global output convergence of a class of continuous-time recurrent neural networks with time-varying thresholds

Author

Liu, Derong ; Hu, Sanqing ; Wang, Jun

Author_Institution

Dept. of Electr. & Comput. Eng., Univ. of Illinois, Chicago, IL, USA

Volume

51

Issue

4

fYear

2004

fDate

4/1/2004 12:00:00 AM

Firstpage

161

Lastpage

167

Abstract

This paper discusses the global output convergence of a class of continuous-time recurrent neural networks (RNNs) with globally Lipschitz continuous and monotone nondecreasing activation functions and locally Lipschitz continuous time-varying thresholds. We establish one sufficient condition to guarantee the global output convergence of this class of neural networks. The present result does not require symmetry in the connection weight matrix. The convergence result is useful in the design of recurrent neural networks with time-varying thresholds.

Keywords

continuous time systems; convergence; recurrent neural nets; time-varying systems; transfer functions; Lipschitz continuity; Lyapunov diagonal semistability; connection weight matrix; continuous-time RNN; global output convergence; globally Lipschitz continuous function; locally Lipschitz continuous time-varying thresholds; monotone nondecreasing activation function; neural network class; recurrent neural networks; time-varying threshold; Asymptotic stability; CADCAM; Computer aided manufacturing; Convergence; Hopfield neural networks; Linear programming; Neural networks; Recurrent neural networks; Sufficient conditions; Vectors;

fLanguage

English

Journal_Title

Circuits and Systems II: Express Briefs, IEEE Transactions on

Publisher

ieee

ISSN

1549-7747

Type

jour

DOI

10.1109/TCSII.2004.824041

Filename

1288419