A Divide-And-Conquer Fast Implementation of Radial Basis Function Networks with Application to Time Series Forecasting

Rong Bo Hunag; Yiu Ming Cheung; Lap Tak Law

doi:10.1142/9789812704955_0008

A Divide-And-Conquer Fast Implementation of Radial Basis Function Networks with Application to Time Series Forecasting

Rong Bo Hunag, Yiu Ming Cheung, Lap Tak Law

Department of Computer Science

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

Abstract

From the dual structural radial basis function network (DSRBF) (Cheung and Xu 2001), this paper presents a new divide-and-conquer learning approach to radial basis function networks (DCRBF). The DCRBF network is a hybrid system consisting of several sub-RBF networks, each of which takes a sub-input space as its input. Since this system divides a high-dimensional modeling problem into serveral low-dimensional ones, it can considerably reduce the structural complexity of a RBF network, whereby the net’s learning is much faster. We have experimentally shown its outstanding learning performance on forecasting two real time series as well as synthetic data in comparison with a conventional RBF one.

Original language	English
Title of host publication	Advances in Data Mining and Modeling
Subtitle of host publication	Advances in Data Mining and Modeling, Hong Kong, 27 – 28 June 2002
Editors	Wai Ki Ching, Michael Kwok Po Ng
Place of Publication	Singapore
Publisher	World Scientific Publishing Co. Pte Ltd
Pages	97-106
Number of pages	10
ISBN (Electronic)	9789814486118
ISBN (Print)	9789812383549
DOIs	https://doi.org/10.1142/9789812704955_0008
Publication status	Published - Apr 2003

Access to Document

10.1142/9789812704955_0008

Cite this

Hunag, R. B., Cheung, Y. M., & Law, L. T. (2003). A Divide-And-Conquer Fast Implementation of Radial Basis Function Networks with Application to Time Series Forecasting. In W. K. Ching, & M. K. P. Ng (Eds.), Advances in Data Mining and Modeling: Advances in Data Mining and Modeling, Hong Kong, 27 – 28 June 2002 (pp. 97-106). World Scientific Publishing Co. Pte Ltd. https://doi.org/10.1142/9789812704955_0008

Hunag, Rong Bo ; Cheung, Yiu Ming ; Law, Lap Tak. / A Divide-And-Conquer Fast Implementation of Radial Basis Function Networks with Application to Time Series Forecasting. Advances in Data Mining and Modeling: Advances in Data Mining and Modeling, Hong Kong, 27 – 28 June 2002. editor / Wai Ki Ching ; Michael Kwok Po Ng. Singapore : World Scientific Publishing Co. Pte Ltd, 2003. pp. 97-106

@inproceedings{849087a57cc841f4a1a2befc8099cd81,

title = "A Divide-And-Conquer Fast Implementation of Radial Basis Function Networks with Application to Time Series Forecasting",

abstract = "From the dual structural radial basis function network (DSRBF) (Cheung and Xu 2001), this paper presents a new divide-and-conquer learning approach to radial basis function networks (DCRBF). The DCRBF network is a hybrid system consisting of several sub-RBF networks, each of which takes a sub-input space as its input. Since this system divides a high-dimensional modeling problem into serveral low-dimensional ones, it can considerably reduce the structural complexity of a RBF network, whereby the net{\textquoteright}s learning is much faster. We have experimentally shown its outstanding learning performance on forecasting two real time series as well as synthetic data in comparison with a conventional RBF one.",

author = "Hunag, {Rong Bo} and Cheung, {Yiu Ming} and Law, {Lap Tak}",

note = "Funding information: The work described in this paper was fully supported by a Faculty Research Grant of Hong Kong Baptist University with Project Number: FRG/02-03/I-06. Publisher copyright: {\textcopyright} 2003 by World Scientific Publishing Co. Re. Ltd.",

year = "2003",

month = apr,

doi = "10.1142/9789812704955_0008",

language = "English",

isbn = "9789812383549",

pages = "97--106",

editor = "Ching, {Wai Ki} and Ng, {Michael Kwok Po}",

booktitle = "Advances in Data Mining and Modeling",

publisher = "World Scientific Publishing Co. Pte Ltd",

address = "Singapore",

}

Hunag, RB, Cheung, YM & Law, LT 2003, A Divide-And-Conquer Fast Implementation of Radial Basis Function Networks with Application to Time Series Forecasting. in WK Ching & MKP Ng (eds), Advances in Data Mining and Modeling: Advances in Data Mining and Modeling, Hong Kong, 27 – 28 June 2002. World Scientific Publishing Co. Pte Ltd, Singapore, pp. 97-106. https://doi.org/10.1142/9789812704955_0008

A Divide-And-Conquer Fast Implementation of Radial Basis Function Networks with Application to Time Series Forecasting. / Hunag, Rong Bo; Cheung, Yiu Ming; Law, Lap Tak.
Advances in Data Mining and Modeling: Advances in Data Mining and Modeling, Hong Kong, 27 – 28 June 2002. ed. / Wai Ki Ching; Michael Kwok Po Ng. Singapore: World Scientific Publishing Co. Pte Ltd, 2003. p. 97-106.

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

TY - GEN

T1 - A Divide-And-Conquer Fast Implementation of Radial Basis Function Networks with Application to Time Series Forecasting

AU - Hunag, Rong Bo

AU - Cheung, Yiu Ming

AU - Law, Lap Tak

N1 - Funding information: The work described in this paper was fully supported by a Faculty Research Grant of Hong Kong Baptist University with Project Number: FRG/02-03/I-06. Publisher copyright: © 2003 by World Scientific Publishing Co. Re. Ltd.

PY - 2003/4

Y1 - 2003/4

N2 - From the dual structural radial basis function network (DSRBF) (Cheung and Xu 2001), this paper presents a new divide-and-conquer learning approach to radial basis function networks (DCRBF). The DCRBF network is a hybrid system consisting of several sub-RBF networks, each of which takes a sub-input space as its input. Since this system divides a high-dimensional modeling problem into serveral low-dimensional ones, it can considerably reduce the structural complexity of a RBF network, whereby the net’s learning is much faster. We have experimentally shown its outstanding learning performance on forecasting two real time series as well as synthetic data in comparison with a conventional RBF one.

AB - From the dual structural radial basis function network (DSRBF) (Cheung and Xu 2001), this paper presents a new divide-and-conquer learning approach to radial basis function networks (DCRBF). The DCRBF network is a hybrid system consisting of several sub-RBF networks, each of which takes a sub-input space as its input. Since this system divides a high-dimensional modeling problem into serveral low-dimensional ones, it can considerably reduce the structural complexity of a RBF network, whereby the net’s learning is much faster. We have experimentally shown its outstanding learning performance on forecasting two real time series as well as synthetic data in comparison with a conventional RBF one.

UR - https://doi.org/10.1142/5260

U2 - 10.1142/9789812704955_0008

DO - 10.1142/9789812704955_0008

M3 - Conference proceeding

SN - 9789812383549

SP - 97

EP - 106

BT - Advances in Data Mining and Modeling

A2 - Ching, Wai Ki

A2 - Ng, Michael Kwok Po

PB - World Scientific Publishing Co. Pte Ltd

CY - Singapore

ER -

Hunag RB, Cheung YM, Law LT. A Divide-And-Conquer Fast Implementation of Radial Basis Function Networks with Application to Time Series Forecasting. In Ching WK, Ng MKP, editors, Advances in Data Mining and Modeling: Advances in Data Mining and Modeling, Hong Kong, 27 – 28 June 2002. Singapore: World Scientific Publishing Co. Pte Ltd. 2003. p. 97-106 doi: 10.1142/9789812704955_0008

A Divide-And-Conquer Fast Implementation of Radial Basis Function Networks with Application to Time Series Forecasting

Abstract

Access to Document

Other files and links

Fingerprint

Cite this