Introduction to the theory and applications of neural networks with quadratic junctions

Nicholas DeClaris, Mu Chun Su

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

7 Scopus citations

Abstract

This paper provides a statistical viewpoint for understanding and using a novel class of neural networks that contain quadratic junctions. It is shown that any Gaussian classifier can be mapped into a quadratic neuron. When the data clusters by means of hyperellipsoids the quadratic neurons provide significant advantages over other representation schemes. Moreover there are cases that even when the data are non-Gaussian, the multilayer neural networks composed of quadratic neurons provide efficient solutions to these pattern recognition problems.

Original languageEnglish
Title of host publication1992 IEEE International Conference on Systems, Man, and Cybernetics
Subtitle of host publicationEmergent Innovations in Information Transfer Processing and Decision Making, SMC 1992
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1320-1325
Number of pages6
ISBN (Electronic)0780307208, 9780780307209
DOIs
StatePublished - 1992
EventIEEE International Conference on Systems, Man, and Cybernetics, SMC 1992 - Chicago, United States
Duration: 18 Oct 199221 Oct 1992

Publication series

NameConference Proceedings - IEEE International Conference on Systems, Man and Cybernetics
Volume1992-January
ISSN (Print)1062-922X

Conference

ConferenceIEEE International Conference on Systems, Man, and Cybernetics, SMC 1992
Country/TerritoryUnited States
CityChicago
Period18/10/9221/10/92

Fingerprint

Dive into the research topics of 'Introduction to the theory and applications of neural networks with quadratic junctions'. Together they form a unique fingerprint.

Cite this