Inhibition and the output map of MVQ neural networks

Author

Abouali, A.H. ; Porter, W.A.

Author_Institution

Egyptian Res. Center, Cairo, Egypt

fYear

1998

fDate

8-10 Mar 1998

Firstpage

397

Lastpage

401

Abstract

We complete the design process of the multiple class vector quantization (MVQ) neural network. The focus of this study is the design of the output layer. The output layer is a group of significant neurons sets. Each set has its own mapping function to the output. The inhibition function enables a single mapping to the output. We also present the canonical architecture of the MVQ network that allows a mix of neurons and the use of higher order moments

Keywords

learning (artificial intelligence); multilayer perceptrons; neural net architecture; MVQ neural networks; canonical architecture; higher order moments; inhibition function; mapping function; multiple class vector quantization neural network; output map; Books; Communication channels; Displays; Neural networks; Neurons; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

System Theory, 1998. Proceedings of the Thirtieth Southeastern Symposium on

Conference_Location

Morgantown, WV

ISSN

0094-2898

Print_ISBN

0-7803-4547-9

Type

conf

DOI

10.1109/SSST.1998.660104

Filename

660104

Link To Document

https://search.isc.ac/dl/search/defaultta.aspx?DTC=49&DC=3069350