Uncertainty propagation through radial basis function networks part II: classification networks

S.G. Pierce, K. Worden, G. Manson

Research output: Contribution to conferencePaper

Abstract

Radial basis function neural networks were trained using both partially supervised and fully supervised training techniques on a simple two-dimensional Gaussian data set. Forward uncertainty propagation through these networks was assessed using a technique of nested interval sets to form an information-gap model of classification performance of the networks. We demonstrate that the interval technique allows both the quantification of worst case and best case error performance of an individual network; and additionally provides an effective tool for optimal network selection in the presence of uncertainty.
LanguageEnglish
Pages929-934
Number of pages5
Publication statusPublished - 2005

Fingerprint

Radial basis function networks
Neural networks
Uncertainty

Keywords

  • Radial Basis Function
  • artificial neural network
  • Ben-Haim’s concept of
  • multi-layer perceptron network

Cite this

@conference{2f42634a85cd41b7b092af0a4eed9535,
title = "Uncertainty propagation through radial basis function networks part II: classification networks",
abstract = "Radial basis function neural networks were trained using both partially supervised and fully supervised training techniques on a simple two-dimensional Gaussian data set. Forward uncertainty propagation through these networks was assessed using a technique of nested interval sets to form an information-gap model of classification performance of the networks. We demonstrate that the interval technique allows both the quantification of worst case and best case error performance of an individual network; and additionally provides an effective tool for optimal network selection in the presence of uncertainty.",
keywords = "Radial Basis Function, artificial neural network, Ben-Haim’s concept of, multi-layer perceptron network",
author = "S.G. Pierce and K. Worden and G. Manson",
note = "Requires Template change to Chapter in Book/Report/Conference proceeding › Conference contribution",
year = "2005",
language = "English",
pages = "929--934",

}

Uncertainty propagation through radial basis function networks part II: classification networks. / Pierce, S.G.; Worden, K.; Manson, G.

2005. 929-934.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Uncertainty propagation through radial basis function networks part II: classification networks

AU - Pierce, S.G.

AU - Worden, K.

AU - Manson, G.

N1 - Requires Template change to Chapter in Book/Report/Conference proceeding › Conference contribution

PY - 2005

Y1 - 2005

N2 - Radial basis function neural networks were trained using both partially supervised and fully supervised training techniques on a simple two-dimensional Gaussian data set. Forward uncertainty propagation through these networks was assessed using a technique of nested interval sets to form an information-gap model of classification performance of the networks. We demonstrate that the interval technique allows both the quantification of worst case and best case error performance of an individual network; and additionally provides an effective tool for optimal network selection in the presence of uncertainty.

AB - Radial basis function neural networks were trained using both partially supervised and fully supervised training techniques on a simple two-dimensional Gaussian data set. Forward uncertainty propagation through these networks was assessed using a technique of nested interval sets to form an information-gap model of classification performance of the networks. We demonstrate that the interval technique allows both the quantification of worst case and best case error performance of an individual network; and additionally provides an effective tool for optimal network selection in the presence of uncertainty.

KW - Radial Basis Function

KW - artificial neural network

KW - Ben-Haim’s concept of

KW - multi-layer perceptron network

UR - http://www.eurodyn2005.univ-mlv.fr/welcome.htm

UR - http://www.millpress.nl/shop/catalogue%20media/978-90-5966-033-5.pdf

UR - http://www.dist.unina.it/proc/2005/EURODYN/pdf/1580.pdf

M3 - Paper

SP - 929

EP - 934

ER -