A fast hierarchical neural network for pattern classification

Author

Annadurai, S. ; Paulpandian, T. ; Balasubramaniam, A.

Author_Institution

Sch. of Comput. Sci. & Eng., Anna Univ., Madras, India

fYear

1994

fDate

29 Nov-2 Dec 1994

Firstpage

110

Lastpage

114

Abstract

A fast hierarchical neural network pattern classifier is proposed in this paper for pattern classification. It is trained with the multilayer perceptron algorithm. It uses only boundary pattern samples and consists of two stages. The first stage provides an approximate solution whereas the second stage gives the exact solution. The experimental results show that the learning time required for the proposed hierarchical network is much less than the standard multilayer perceptron classifier

Keywords

backpropagation; learning (artificial intelligence); multilayer perceptrons; pattern classification; approximate solution; boundary pattern samples; exact solution; fast hierarchical neural network; hierarchical network; learning time; multilayer perceptron algorithm; multilayer perceptron classifier; neural network pattern classifier; pattern classification; Backpropagation; Computer science; Computer vision; Decision making; Educational institutions; Multi-layer neural network; Multilayer perceptrons; Neural networks; Pattern classification; Signal processing;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Information Systems,1994. Proceedings of the 1994 Second Australian and New Zealand Conference on

Conference_Location

Brisbane, Qld.

Print_ISBN

0-7803-2404-8

Type

conf

DOI

10.1109/ANZIIS.1994.396939

Filename

396939