Volterra-Wiener characterization of a recurrent neural network

N. Z. Hakim, J. J. Kaufman, G. Cerf, H. E. Meadows

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

The Volterra-Wiener theory of nonlinear systems to neural network modeling is studied. A recursive formula to compute the Volterra kernels of a well-known recurrent neural network is presented. Expressing the neural network input-output relationship in this framework allows issues such as generalization, functional representation power, and optimal choice of the neuron activation function to be addressed. The issue of neural network realization of a subclass of systems admitting Volterra expansion is also addressed. Such an approach is expected to shed some light on these issues and help design neural networks as models for signal processing and nonlinear control.

Original languageEnglish
Title of host publicationProceedings of the Annual Conference on Engineering in Medicine and Biology
PublisherPubl by IEEE
Pages1397-1398
Number of pages2
Editionpt 3
ISBN (Print)0780302168
StatePublished - 1991
Externally publishedYes
EventProceedings of the 13th Annual International Conference of the IEEE Engineering in Medicine and Biology Society - Orlando, FL, USA
Duration: 31 Oct 19913 Nov 1991

Publication series

NameProceedings of the Annual Conference on Engineering in Medicine and Biology
Numberpt 3
Volume13
ISSN (Print)0589-1019

Conference

ConferenceProceedings of the 13th Annual International Conference of the IEEE Engineering in Medicine and Biology Society
CityOrlando, FL, USA
Period31/10/913/11/91

Fingerprint

Dive into the research topics of 'Volterra-Wiener characterization of a recurrent neural network'. Together they form a unique fingerprint.

Cite this