Discriminant analysis by a neural network with mahalanobis distance

Yoshifusa Ito, Cidambi Srinivasan, Hiroyuki Izumi

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

12 Scopus citations

Abstract

We propose a neural network which can approximate Mahalanobis discriminant functions after being trained. It can be realized if a Bayesian neural network is equipped with two additional subnetworks. The training is performed sequentially and, hence, the past teacher signals need not be memorized. In this paper, we treat the two-category normal-distribution case. The results of simple simulations are included.

Original languageEnglish
Title of host publicationArtificial Neural Networks, ICANN 2006 - 16th International Conference, Proceedings
Pages350-360
Number of pages11
DOIs
StatePublished - 2006
Event16th International Conference on Artificial Neural Networks, ICANN 2006 - Athens, Greece
Duration: Sep 10 2006Sep 14 2006

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume4132 LNCS - II
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference16th International Conference on Artificial Neural Networks, ICANN 2006
Country/TerritoryGreece
CityAthens
Period9/10/069/14/06

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science (all)

Fingerprint

Dive into the research topics of 'Discriminant analysis by a neural network with mahalanobis distance'. Together they form a unique fingerprint.

Cite this