Study of multi-branch structure of Universal Learning Networks

Shingo Mabu, Kaoru Shimada, Kotaro Hirasawa, Takayuki Furuzuki

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

In this paper, multi-branch structure of Universal Learning Networks (ULNs) is studied to verify its effectiveness for obtaining compact models, which have neurons connected with other neurons using more than two branches having nonlinear functions. Multi-branch structure has been proved to have higher representation/generalization ability and lower computational cost than conventional neural networks because of the nonlinear function of the multi-branches and the reduction of the number of neurons to be used. In addition, learning of delay elements of multi-branch ULNs has improved their potential to build up a compact dynamical model with higher performances and lower computational cost when applied for identifying dynamical systems.

Original languageEnglish
Pages (from-to)393-403
Number of pages11
JournalApplied Soft Computing Journal
Volume9
Issue number1
DOIs
Publication statusPublished - 2009 Jan

Fingerprint

Neurons
Costs
Dynamical systems
Neural networks

Keywords

  • Multi-branch structure
  • Time-delayed recurrent networks
  • Universal Learning Networks

ASJC Scopus subject areas

  • Software

Cite this

Study of multi-branch structure of Universal Learning Networks. / Mabu, Shingo; Shimada, Kaoru; Hirasawa, Kotaro; Furuzuki, Takayuki.

In: Applied Soft Computing Journal, Vol. 9, No. 1, 01.2009, p. 393-403.

Research output: Contribution to journalArticle

Mabu, Shingo ; Shimada, Kaoru ; Hirasawa, Kotaro ; Furuzuki, Takayuki. / Study of multi-branch structure of Universal Learning Networks. In: Applied Soft Computing Journal. 2009 ; Vol. 9, No. 1. pp. 393-403.
@article{5ff53fd1d55e454e85b8868425b6e78b,
title = "Study of multi-branch structure of Universal Learning Networks",
abstract = "In this paper, multi-branch structure of Universal Learning Networks (ULNs) is studied to verify its effectiveness for obtaining compact models, which have neurons connected with other neurons using more than two branches having nonlinear functions. Multi-branch structure has been proved to have higher representation/generalization ability and lower computational cost than conventional neural networks because of the nonlinear function of the multi-branches and the reduction of the number of neurons to be used. In addition, learning of delay elements of multi-branch ULNs has improved their potential to build up a compact dynamical model with higher performances and lower computational cost when applied for identifying dynamical systems.",
keywords = "Multi-branch structure, Time-delayed recurrent networks, Universal Learning Networks",
author = "Shingo Mabu and Kaoru Shimada and Kotaro Hirasawa and Takayuki Furuzuki",
year = "2009",
month = "1",
doi = "10.1016/j.asoc.2008.01.012",
language = "English",
volume = "9",
pages = "393--403",
journal = "Applied Soft Computing",
issn = "1568-4946",
publisher = "Elsevier BV",
number = "1",

}

TY - JOUR

T1 - Study of multi-branch structure of Universal Learning Networks

AU - Mabu, Shingo

AU - Shimada, Kaoru

AU - Hirasawa, Kotaro

AU - Furuzuki, Takayuki

PY - 2009/1

Y1 - 2009/1

N2 - In this paper, multi-branch structure of Universal Learning Networks (ULNs) is studied to verify its effectiveness for obtaining compact models, which have neurons connected with other neurons using more than two branches having nonlinear functions. Multi-branch structure has been proved to have higher representation/generalization ability and lower computational cost than conventional neural networks because of the nonlinear function of the multi-branches and the reduction of the number of neurons to be used. In addition, learning of delay elements of multi-branch ULNs has improved their potential to build up a compact dynamical model with higher performances and lower computational cost when applied for identifying dynamical systems.

AB - In this paper, multi-branch structure of Universal Learning Networks (ULNs) is studied to verify its effectiveness for obtaining compact models, which have neurons connected with other neurons using more than two branches having nonlinear functions. Multi-branch structure has been proved to have higher representation/generalization ability and lower computational cost than conventional neural networks because of the nonlinear function of the multi-branches and the reduction of the number of neurons to be used. In addition, learning of delay elements of multi-branch ULNs has improved their potential to build up a compact dynamical model with higher performances and lower computational cost when applied for identifying dynamical systems.

KW - Multi-branch structure

KW - Time-delayed recurrent networks

KW - Universal Learning Networks

UR - http://www.scopus.com/inward/record.url?scp=53749093225&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=53749093225&partnerID=8YFLogxK

U2 - 10.1016/j.asoc.2008.01.012

DO - 10.1016/j.asoc.2008.01.012

M3 - Article

VL - 9

SP - 393

EP - 403

JO - Applied Soft Computing

JF - Applied Soft Computing

SN - 1568-4946

IS - 1

ER -