A fast and efficient competitive learning design algorithm based on weight vector training in transform domain

Wen-Jyi Hwang, Yi Chong Zeng, Shi Chiang Liao

Research output: Contribution to journalArticle

Abstract

This paper presents a new competitive learning (CL) algorithm which performs the training in the wavelet domain. In the algorithm, the winning neural units during the training process are identified using the partial distance search (PDS) technique so that little multiplication is required. The PDS can be performed over the lower resolution representation of codewords in the wavelet transform domain to further reduce the computation time required for training. Simulation results show that, at the expense of a possible slight decrease in performance, the algorithm requires less than 5% of the computational time required by the traditional CL algorithm in many cases.

Original languageEnglish
Pages (from-to)625-631
Number of pages7
JournalJournal of the Chinese Institute of Engineers, Transactions of the Chinese Institute of Engineers,Series A/Chung-kuo Kung Ch'eng Hsuch K'an
Volume21
Issue number5
DOIs
Publication statusPublished - 1998 Jan 1

Fingerprint

Learning algorithms
Wavelet transforms

Keywords

  • Competitive Learning
  • Neural Networks
  • Wavelet Transform

ASJC Scopus subject areas

  • Engineering(all)

Cite this

@article{4a19176f94544af1921fcb94cf7ab32a,
title = "A fast and efficient competitive learning design algorithm based on weight vector training in transform domain",
abstract = "This paper presents a new competitive learning (CL) algorithm which performs the training in the wavelet domain. In the algorithm, the winning neural units during the training process are identified using the partial distance search (PDS) technique so that little multiplication is required. The PDS can be performed over the lower resolution representation of codewords in the wavelet transform domain to further reduce the computation time required for training. Simulation results show that, at the expense of a possible slight decrease in performance, the algorithm requires less than 5{\%} of the computational time required by the traditional CL algorithm in many cases.",
keywords = "Competitive Learning, Neural Networks, Wavelet Transform",
author = "Wen-Jyi Hwang and Zeng, {Yi Chong} and Liao, {Shi Chiang}",
year = "1998",
month = "1",
day = "1",
doi = "10.1080/02533839.1998.9670424",
language = "English",
volume = "21",
pages = "625--631",
journal = "Chung-kuo Kung Ch'eng Hsueh K'an/Journal of the Chinese Institute of Engineers",
issn = "0253-3839",
publisher = "Chinese Institute of Engineers",
number = "5",

}

TY - JOUR

T1 - A fast and efficient competitive learning design algorithm based on weight vector training in transform domain

AU - Hwang, Wen-Jyi

AU - Zeng, Yi Chong

AU - Liao, Shi Chiang

PY - 1998/1/1

Y1 - 1998/1/1

N2 - This paper presents a new competitive learning (CL) algorithm which performs the training in the wavelet domain. In the algorithm, the winning neural units during the training process are identified using the partial distance search (PDS) technique so that little multiplication is required. The PDS can be performed over the lower resolution representation of codewords in the wavelet transform domain to further reduce the computation time required for training. Simulation results show that, at the expense of a possible slight decrease in performance, the algorithm requires less than 5% of the computational time required by the traditional CL algorithm in many cases.

AB - This paper presents a new competitive learning (CL) algorithm which performs the training in the wavelet domain. In the algorithm, the winning neural units during the training process are identified using the partial distance search (PDS) technique so that little multiplication is required. The PDS can be performed over the lower resolution representation of codewords in the wavelet transform domain to further reduce the computation time required for training. Simulation results show that, at the expense of a possible slight decrease in performance, the algorithm requires less than 5% of the computational time required by the traditional CL algorithm in many cases.

KW - Competitive Learning

KW - Neural Networks

KW - Wavelet Transform

UR - http://www.scopus.com/inward/record.url?scp=0032166635&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032166635&partnerID=8YFLogxK

U2 - 10.1080/02533839.1998.9670424

DO - 10.1080/02533839.1998.9670424

M3 - Article

AN - SCOPUS:0032166635

VL - 21

SP - 625

EP - 631

JO - Chung-kuo Kung Ch'eng Hsueh K'an/Journal of the Chinese Institute of Engineers

JF - Chung-kuo Kung Ch'eng Hsueh K'an/Journal of the Chinese Institute of Engineers

SN - 0253-3839

IS - 5

ER -