A global transformation approach to RBF neural network learning

Kar Ann Toh, K. Z. Mao

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

In this paper, we propose to train the RBF neural network using a global descent method. Essentially, the method imposes a monotonic transformation on the training objective to improve numerical sensitivity without altering the relative orders of all local extrema. A gradient descent search which inherits the global descent property is derived to locate the global solution of an error objective. Numerical examples comparing the global descent algorithm with a gradient-based line-search algorithm shows superiority of the proposed global descent algorithm in terms of speed of convergence and quality of solution achieved.

Original languageEnglish
Pages (from-to)96-99
Number of pages4
JournalProceedings - International Conference on Pattern Recognition
Volume16
Issue number2
Publication statusPublished - 2002 Dec 1

Fingerprint

Neural networks

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition

Cite this

@article{288591c7922c4db9972d0311665c117a,
title = "A global transformation approach to RBF neural network learning",
abstract = "In this paper, we propose to train the RBF neural network using a global descent method. Essentially, the method imposes a monotonic transformation on the training objective to improve numerical sensitivity without altering the relative orders of all local extrema. A gradient descent search which inherits the global descent property is derived to locate the global solution of an error objective. Numerical examples comparing the global descent algorithm with a gradient-based line-search algorithm shows superiority of the proposed global descent algorithm in terms of speed of convergence and quality of solution achieved.",
author = "Toh, {Kar Ann} and Mao, {K. Z.}",
year = "2002",
month = "12",
day = "1",
language = "English",
volume = "16",
pages = "96--99",
journal = "Proceedings - International Conference on Pattern Recognition",
issn = "1051-4651",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "2",

}

A global transformation approach to RBF neural network learning. / Toh, Kar Ann; Mao, K. Z.

In: Proceedings - International Conference on Pattern Recognition, Vol. 16, No. 2, 01.12.2002, p. 96-99.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A global transformation approach to RBF neural network learning

AU - Toh, Kar Ann

AU - Mao, K. Z.

PY - 2002/12/1

Y1 - 2002/12/1

N2 - In this paper, we propose to train the RBF neural network using a global descent method. Essentially, the method imposes a monotonic transformation on the training objective to improve numerical sensitivity without altering the relative orders of all local extrema. A gradient descent search which inherits the global descent property is derived to locate the global solution of an error objective. Numerical examples comparing the global descent algorithm with a gradient-based line-search algorithm shows superiority of the proposed global descent algorithm in terms of speed of convergence and quality of solution achieved.

AB - In this paper, we propose to train the RBF neural network using a global descent method. Essentially, the method imposes a monotonic transformation on the training objective to improve numerical sensitivity without altering the relative orders of all local extrema. A gradient descent search which inherits the global descent property is derived to locate the global solution of an error objective. Numerical examples comparing the global descent algorithm with a gradient-based line-search algorithm shows superiority of the proposed global descent algorithm in terms of speed of convergence and quality of solution achieved.

UR - http://www.scopus.com/inward/record.url?scp=33751557924&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33751557924&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:33751557924

VL - 16

SP - 96

EP - 99

JO - Proceedings - International Conference on Pattern Recognition

JF - Proceedings - International Conference on Pattern Recognition

SN - 1051-4651

IS - 2

ER -