The augmented ART1 neural network

Author

Heileman, Gregory L. ; Georgiopoulos, Michael

Author_Institution

Dept. of Electr. & Comput. Eng., New Mexico Univ., Albuquerque, NM, USA

fYear

1991

fDate

8-14 Jul 1991

Firstpage

467

Abstract

A real-time neural network AART 1 (augmented adaptive resonance theory), that describes the dynamics of the ART 1 model is presented. This involves modifying the architecture of the ART 1 network through the addition of a reset node and a set of inhibitory nodes in the F2 layer. Along with modifications to the differential equations describing the F1 and F2 layer nodes, these changes embody the functionality of the orienting subsystem in the ART 1 model. In addition, the modifications described allow the ART 1 model to be implemented solely using a set of concurrently executing nonlinear differential equations. Thus, the AART1 network requires no external control features. Computer simulation results demonstrate the efficacy of this network

Keywords

adaptive systems; neural nets; nonlinear differential equations; real-time systems; resonance; virtual machines; F2 layer; adaptive resonance theory; architectural modifications; augmented ART1 neural network; computer simulation; concurrently executing nonlinear differential equations; inhibitory nodes; model dynamics; orienting subsystem; real-time neural network; reset node; Algorithm design and analysis; Circuits; Differential equations; Gain control; Neural networks; Nonlinear equations; Out of order; Real time systems; Testing; Time factors;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks, 1991., IJCNN-91-Seattle International Joint Conference on

Conference_Location

Seattle, WA

Print_ISBN

0-7803-0164-1

Type

conf

DOI

10.1109/IJCNN.1991.155378

Filename

155378