Multicategory Bayesian decision using a three-layer neural network

Yoshifusa Ito, Cidambi Srinivasan

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

10 Scopus citations

Abstract

We realize a multicategory Bayesian classifier by a three-layer neural network having rather a small number of hidden layer units. The state-conditional probability distributions are supposed to be multivariate normal distributions. The network has direct connections between the input and output layers. Its outputs are monotone mappings of posterior probabilities. Hence, they can be used as discriminant functions and, in addition, the posterior probabilities can be easily retrieved from the outputs.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
EditorsOkyay Kaynak, Ethem Alpaydin, Erkki Oja, Lei Xu
Pages253-261
Number of pages9
DOIs
StatePublished - 2003

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2714
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science (all)

Fingerprint

Dive into the research topics of 'Multicategory Bayesian decision using a three-layer neural network'. Together they form a unique fingerprint.

Cite this