A functions localized neural network with branch gates

Qingyu Xiong; Kotaro Hirasawa; Jinglu Hu; Junichi Murata

doi:10.1016/S0893-6080(03)00211-9

A functions localized neural network with branch gates

Qingyu Xiong, Kotaro Hirasawa^*, Jinglu Hu, Junichi Murata

^*Corresponding author for this work

Graduate School of Information, Production and Systems

Research output: Contribution to journal › Article › peer-review

9 Citations (Scopus)

Abstract

In this paper, a functions localized network with branch gates (FLN-bg) is studied, which consists of a basic network and a branch gate network. The branch gate network is used to determine which intermediate nodes of the basic network should be connected to the output node with a gate coefficient ranging from 0 to 1. This determination will adjust the outputs of the intermediate nodes of the basic network depending on the values of the inputs of the network in order to realize a functions localized network. FLN-bg is applied to function approximation problems and a two-spiral problem. The simulation results show that FLN-bg exhibits better performance than conventional neural networks with comparable complexity.

Original language	English
Pages (from-to)	1461-1481
Number of pages	21
Journal	Neural Networks
Volume	16
Issue number	10
DOIs	https://doi.org/10.1016/S0893-6080(03)00211-9
Publication status	Published - 2003 Dec

Keywords

Branch gate
Functions localization
Fuzzy networks
Neural networks
Universal learning networks

ASJC Scopus subject areas

Cognitive Neuroscience
Artificial Intelligence

Access to Document

10.1016/S0893-6080(03)00211-9

Cite this

@article{462ee6cf5c334accb3f8971e72456ea4,

title = "A functions localized neural network with branch gates",

abstract = "In this paper, a functions localized network with branch gates (FLN-bg) is studied, which consists of a basic network and a branch gate network. The branch gate network is used to determine which intermediate nodes of the basic network should be connected to the output node with a gate coefficient ranging from 0 to 1. This determination will adjust the outputs of the intermediate nodes of the basic network depending on the values of the inputs of the network in order to realize a functions localized network. FLN-bg is applied to function approximation problems and a two-spiral problem. The simulation results show that FLN-bg exhibits better performance than conventional neural networks with comparable complexity.",

keywords = "Branch gate, Functions localization, Fuzzy networks, Neural networks, Universal learning networks",

author = "Qingyu Xiong and Kotaro Hirasawa and Jinglu Hu and Junichi Murata",

year = "2003",

month = dec,

doi = "10.1016/S0893-6080(03)00211-9",

language = "English",

volume = "16",

pages = "1461--1481",

journal = "Neural Networks",

issn = "0893-6080",

publisher = "Elsevier Limited",

number = "10",

}

TY - JOUR

T1 - A functions localized neural network with branch gates

AU - Xiong, Qingyu

AU - Hirasawa, Kotaro

AU - Hu, Jinglu

AU - Murata, Junichi

PY - 2003/12

Y1 - 2003/12

N2 - In this paper, a functions localized network with branch gates (FLN-bg) is studied, which consists of a basic network and a branch gate network. The branch gate network is used to determine which intermediate nodes of the basic network should be connected to the output node with a gate coefficient ranging from 0 to 1. This determination will adjust the outputs of the intermediate nodes of the basic network depending on the values of the inputs of the network in order to realize a functions localized network. FLN-bg is applied to function approximation problems and a two-spiral problem. The simulation results show that FLN-bg exhibits better performance than conventional neural networks with comparable complexity.

AB - In this paper, a functions localized network with branch gates (FLN-bg) is studied, which consists of a basic network and a branch gate network. The branch gate network is used to determine which intermediate nodes of the basic network should be connected to the output node with a gate coefficient ranging from 0 to 1. This determination will adjust the outputs of the intermediate nodes of the basic network depending on the values of the inputs of the network in order to realize a functions localized network. FLN-bg is applied to function approximation problems and a two-spiral problem. The simulation results show that FLN-bg exhibits better performance than conventional neural networks with comparable complexity.

KW - Branch gate

KW - Functions localization

KW - Fuzzy networks

KW - Neural networks

KW - Universal learning networks

UR - http://www.scopus.com/inward/record.url?scp=0242578092&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0242578092&partnerID=8YFLogxK

U2 - 10.1016/S0893-6080(03)00211-9

DO - 10.1016/S0893-6080(03)00211-9

M3 - Article

C2 - 14622877

AN - SCOPUS:0242578092

SN - 0893-6080

VL - 16

SP - 1461

EP - 1481

JO - Neural Networks

JF - Neural Networks

IS - 10

ER -

A functions localized neural network with branch gates

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this