Evaluation of multi-layered RBF networks

Kotaro Hirasawa, Takuya Matsuoka, Masanao Ohbayashi, Junichi Murata

Research output: Contribution to journalConference article

4 Citations (Scopus)

Abstract

In this paper, an investigation into the performance of multi-layered Radial Basis Functions(RBF) networks is conducted which use Gaussian function in place of sigmoidal function in multi-layered Neural Networks(NNs). The focus is on the difference of approximation abilities between multi-layered RBF networks and NNs. A function approximation problem is employed to evaluate the performance of multi-layered RBF networks, and several types of different functions are used as the functions to be approximated. Gradient method is employed to optimize the parameters including centers, widths, and linear connection weights to the output nodes. It is shown from the result that RBF does not always have significant advantages over sigmoidal functions when they are used in multi-layered networks.

Original languageEnglish
Pages (from-to)908-911
Number of pages4
JournalProceedings of the IEEE International Conference on Systems, Man and Cybernetics
Volume1
Publication statusPublished - Dec 1 1997
EventProceedings of the 1997 IEEE International Conference on Systems, Man, and Cybernetics. Part 1 (of 5) - Orlando, FL, USA
Duration: Oct 12 1997Oct 15 1997

Fingerprint

Radial basis function networks
Neural networks
Gradient methods
parameter
method
evaluation

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Hardware and Architecture

Cite this

Evaluation of multi-layered RBF networks. / Hirasawa, Kotaro; Matsuoka, Takuya; Ohbayashi, Masanao; Murata, Junichi.

In: Proceedings of the IEEE International Conference on Systems, Man and Cybernetics, Vol. 1, 01.12.1997, p. 908-911.

Research output: Contribution to journalConference article

Hirasawa, Kotaro ; Matsuoka, Takuya ; Ohbayashi, Masanao ; Murata, Junichi. / Evaluation of multi-layered RBF networks. In: Proceedings of the IEEE International Conference on Systems, Man and Cybernetics. 1997 ; Vol. 1. pp. 908-911.
@article{ae2203bd886743808f67bfd01975fc8f,
title = "Evaluation of multi-layered RBF networks",
abstract = "In this paper, an investigation into the performance of multi-layered Radial Basis Functions(RBF) networks is conducted which use Gaussian function in place of sigmoidal function in multi-layered Neural Networks(NNs). The focus is on the difference of approximation abilities between multi-layered RBF networks and NNs. A function approximation problem is employed to evaluate the performance of multi-layered RBF networks, and several types of different functions are used as the functions to be approximated. Gradient method is employed to optimize the parameters including centers, widths, and linear connection weights to the output nodes. It is shown from the result that RBF does not always have significant advantages over sigmoidal functions when they are used in multi-layered networks.",
author = "Kotaro Hirasawa and Takuya Matsuoka and Masanao Ohbayashi and Junichi Murata",
year = "1997",
month = "12",
day = "1",
language = "English",
volume = "1",
pages = "908--911",
journal = "Proceedings of the IEEE International Conference on Systems, Man and Cybernetics",
issn = "0884-3627",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - Evaluation of multi-layered RBF networks

AU - Hirasawa, Kotaro

AU - Matsuoka, Takuya

AU - Ohbayashi, Masanao

AU - Murata, Junichi

PY - 1997/12/1

Y1 - 1997/12/1

N2 - In this paper, an investigation into the performance of multi-layered Radial Basis Functions(RBF) networks is conducted which use Gaussian function in place of sigmoidal function in multi-layered Neural Networks(NNs). The focus is on the difference of approximation abilities between multi-layered RBF networks and NNs. A function approximation problem is employed to evaluate the performance of multi-layered RBF networks, and several types of different functions are used as the functions to be approximated. Gradient method is employed to optimize the parameters including centers, widths, and linear connection weights to the output nodes. It is shown from the result that RBF does not always have significant advantages over sigmoidal functions when they are used in multi-layered networks.

AB - In this paper, an investigation into the performance of multi-layered Radial Basis Functions(RBF) networks is conducted which use Gaussian function in place of sigmoidal function in multi-layered Neural Networks(NNs). The focus is on the difference of approximation abilities between multi-layered RBF networks and NNs. A function approximation problem is employed to evaluate the performance of multi-layered RBF networks, and several types of different functions are used as the functions to be approximated. Gradient method is employed to optimize the parameters including centers, widths, and linear connection weights to the output nodes. It is shown from the result that RBF does not always have significant advantages over sigmoidal functions when they are used in multi-layered networks.

UR - http://www.scopus.com/inward/record.url?scp=0031380686&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031380686&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:0031380686

VL - 1

SP - 908

EP - 911

JO - Proceedings of the IEEE International Conference on Systems, Man and Cybernetics

JF - Proceedings of the IEEE International Conference on Systems, Man and Cybernetics

SN - 0884-3627

ER -