Title

Gap-based estimation: Choosing the smoothing parameters for Probabilistic and general regression neural networks

Authors

Authors

M. Zhong; D. Coggeshall; E. Ghaneie; T. Pope; M. Rivera; M. Georgiopoulos; G. C. Anagnostopoulos; M. Mollaghasemi;S. Richie

Comments

Authors: contact us about adding a copy of your work at STARS@ucf.edu

Abbreviated Journal Title

Neural Comput.

Keywords

DENSITY; CLASSIFICATION; Computer Science, Artificial Intelligence

Abstract

Probabilistic neural networks (PNN) and general regression neural networks (GRNN) represent knowledge by simple but interpretable models that approximate the optimal classifier or predictor in the sense of expected value of the accuracy. These models require the specification of an important smoothing parameter, which is usually chosen by crossvalidation or clustering. In this letter, we demonstrate the problems with the cross-validation and clustering approaches to specify the smoothing parameter, discuss the relationship between this parameter and some of the data statistics, and attempt to develop a fast approach to determine the optimal value of this parameter. Finally, through experimentation, we show that our approach, referred to as a gap-based estimation approach, is superior in speed to the compared approaches, including support vector machine, and yields good and stable accuracy.

Journal Title

Neural Computation

Volume

19

Issue/Number

10

Publication Date

1-1-2007

Document Type

Article

Language

English

First Page

2840

Last Page

2864

WOS Identifier

WOS:000249243800012

ISSN

0899-7667

Share

COinS