Weighted Parzen windows for radial basis function network design

Author

Babich, Gregory A. ; Sibul, Leon H.

Author_Institution

Appl. Res. Lab., State College, PA, USA

Volume

2

fYear

1994

fDate

31 Oct-2 Nov 1994

Firstpage

897

Abstract

This paper shows how the weighted Parzen window (WPW) technique can be used for radial basis function network (RBFN) design. The WPW training algorithm uses an agglomerative hierarchical clustering procedure to find the RBFN centers and weights. This approach reduces storage requirements as it selects the centers and weights. It is shown that RBFNs can be designed using the WPW technique so that they are functionally equivalent to some statistical techniques. Experimental results are reported for two practical applications, laser-weld classification and handwritten character recognition. The results show that WPW designed RBFNs outperform some neural techniques in these applications

Keywords

character recognition; feedforward neural nets; handwriting recognition; laser beam welding; learning (artificial intelligence); pattern classification; statistical analysis; RBFN; agglomerative hierarchical clustering; experimental results; handwritten character recognition; laser-weld classification; pattern classification; radial basis function network design; statistical techniques; storage requirements reduction; training algorithm; weighted Parzen windows; Artificial neural networks; Character recognition; Clustering algorithms; Density functional theory; Educational institutions; Feedforward systems; Laboratories; Laser applications; Pattern classification; Radial basis function networks;

fLanguage

English

Publisher

ieee

Conference_Titel

Signals, Systems and Computers, 1994. 1994 Conference Record of the Twenty-Eighth Asilomar Conference on

Conference_Location

Pacific Grove, CA

ISSN

1058-6393

Print_ISBN

0-8186-6405-3

Type

conf

DOI

10.1109/ACSSC.1994.471590

Filename

471590