An analog Gaussian generator circuit for radial basis function neural networks with non-volatile center storage

Author

Galbraith, John ; Holman, W. Timothy

Author_Institution

Arizona Univ., Tucson, AZ, USA

Volume

1

fYear

1996

fDate

18-21 Aug 1996

Firstpage

43

Abstract

A BiCMOS analog circuit that computes a Gaussian function suitable for use in radial basis function neural networks is proposed. The circuit accepts multiple voltage inputs and stores the basis function center in a set of nonvolatile floating gate memory cells. The design is well suited for adaptive networks that use on-chip learning. Experimental results of the circuit are shown

Keywords

BiCMOS analogue integrated circuits; Gaussian processes; cellular arrays; feedforward neural nets; learning (artificial intelligence); neural chips; BiCMOS analog circuit; adaptive networks; analog Gaussian generator circuit; floating gate memory cells; multiple voltage inputs; nonvolatile center storage; on-chip learning; radial basis function neural networks; Analog circuits; Analog computers; Artificial intelligence; BiCMOS integrated circuits; Computer networks; Neural networks; Nonvolatile memory; Optical computing; Radial basis function networks; Virtual colonoscopy;

fLanguage

English

Publisher

ieee

Conference_Titel

Circuits and Systems, 1996., IEEE 39th Midwest symposium on

Conference_Location

Ames, IA

Print_ISBN

0-7803-3636-4

Type

conf

DOI

10.1109/MWSCAS.1996.594023

Filename

594023