Multibranch structure and its localized property in layered neural networks

Takashi Yamashita, Kotaro Hirasawa, Takayuki Furuzuki

Research output: Contribution to journalArticle

Abstract

Neural networks (NNs) can solve only simple problems if the network size is too small, but increasing the network size is costly in terms of memory space and calculation time. Thus, we have studied how to construct a network structure with high performance and low cost in space and time. One solution is a multibranch structure. Conventional NNs use the single-branch structure for connections, while the multibranch structure has multiple branches between nodes. In this paper, a new method which enables the multibranch NNs to have the localized property is proposed. It is well known that RBF networks have the localized property, which makes it possible to approximate functions faster than sigmoidal NNs. By using the multibranch structure having the localized property of RBF networks, NNs can obtain superior performance while maintaining lower costs in space and time. Simulation results of function approximations and a classification problem are presented to illustrate the effectiveness of multibranch NNs.

Original languageEnglish
Pages (from-to)48-55
Number of pages8
JournalElectrical Engineering in Japan (English translation of Denki Gakkai Ronbunshi)
Volume162
Issue number1
DOIs
Publication statusPublished - 2008 Jan 15

Fingerprint

Neural networks
Radial basis function networks
Costs
Data storage equipment

Keywords

  • Localized property
  • Multibranch structure
  • Neural networks

ASJC Scopus subject areas

  • Energy Engineering and Power Technology
  • Electrical and Electronic Engineering

Cite this

Multibranch structure and its localized property in layered neural networks. / Yamashita, Takashi; Hirasawa, Kotaro; Furuzuki, Takayuki.

In: Electrical Engineering in Japan (English translation of Denki Gakkai Ronbunshi), Vol. 162, No. 1, 15.01.2008, p. 48-55.

Research output: Contribution to journalArticle

@article{0abfa4d6d7a34d48ba18c694d9352735,
title = "Multibranch structure and its localized property in layered neural networks",
abstract = "Neural networks (NNs) can solve only simple problems if the network size is too small, but increasing the network size is costly in terms of memory space and calculation time. Thus, we have studied how to construct a network structure with high performance and low cost in space and time. One solution is a multibranch structure. Conventional NNs use the single-branch structure for connections, while the multibranch structure has multiple branches between nodes. In this paper, a new method which enables the multibranch NNs to have the localized property is proposed. It is well known that RBF networks have the localized property, which makes it possible to approximate functions faster than sigmoidal NNs. By using the multibranch structure having the localized property of RBF networks, NNs can obtain superior performance while maintaining lower costs in space and time. Simulation results of function approximations and a classification problem are presented to illustrate the effectiveness of multibranch NNs.",
keywords = "Localized property, Multibranch structure, Neural networks",
author = "Takashi Yamashita and Kotaro Hirasawa and Takayuki Furuzuki",
year = "2008",
month = "1",
day = "15",
doi = "10.1002/eej.20352",
language = "English",
volume = "162",
pages = "48--55",
journal = "Electrical Engineering in Japan (English translation of Denki Gakkai Ronbunshi)",
issn = "0424-7760",
publisher = "John Wiley and Sons Inc.",
number = "1",

}

TY - JOUR

T1 - Multibranch structure and its localized property in layered neural networks

AU - Yamashita, Takashi

AU - Hirasawa, Kotaro

AU - Furuzuki, Takayuki

PY - 2008/1/15

Y1 - 2008/1/15

N2 - Neural networks (NNs) can solve only simple problems if the network size is too small, but increasing the network size is costly in terms of memory space and calculation time. Thus, we have studied how to construct a network structure with high performance and low cost in space and time. One solution is a multibranch structure. Conventional NNs use the single-branch structure for connections, while the multibranch structure has multiple branches between nodes. In this paper, a new method which enables the multibranch NNs to have the localized property is proposed. It is well known that RBF networks have the localized property, which makes it possible to approximate functions faster than sigmoidal NNs. By using the multibranch structure having the localized property of RBF networks, NNs can obtain superior performance while maintaining lower costs in space and time. Simulation results of function approximations and a classification problem are presented to illustrate the effectiveness of multibranch NNs.

AB - Neural networks (NNs) can solve only simple problems if the network size is too small, but increasing the network size is costly in terms of memory space and calculation time. Thus, we have studied how to construct a network structure with high performance and low cost in space and time. One solution is a multibranch structure. Conventional NNs use the single-branch structure for connections, while the multibranch structure has multiple branches between nodes. In this paper, a new method which enables the multibranch NNs to have the localized property is proposed. It is well known that RBF networks have the localized property, which makes it possible to approximate functions faster than sigmoidal NNs. By using the multibranch structure having the localized property of RBF networks, NNs can obtain superior performance while maintaining lower costs in space and time. Simulation results of function approximations and a classification problem are presented to illustrate the effectiveness of multibranch NNs.

KW - Localized property

KW - Multibranch structure

KW - Neural networks

UR - http://www.scopus.com/inward/record.url?scp=35348938638&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=35348938638&partnerID=8YFLogxK

U2 - 10.1002/eej.20352

DO - 10.1002/eej.20352

M3 - Article

AN - SCOPUS:35348938638

VL - 162

SP - 48

EP - 55

JO - Electrical Engineering in Japan (English translation of Denki Gakkai Ronbunshi)

JF - Electrical Engineering in Japan (English translation of Denki Gakkai Ronbunshi)

SN - 0424-7760

IS - 1

ER -