2 learning of dynamic neural networks

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

This paper proposes an ℒ2 learning law as a new learning method for dynamic neural networks with external disturbance. Based on linear matrix inequality (LMI) formulation, the ℒ2 learning law is presented to not only guarantee asymptotical stability of dynamic neural networks but also reduce the effect of external disturbance to an ℒ2 induced norm constraint. It is shown that the design of the ℒ2 learning law for such neural networks can be achieved by solving LMIs, which can be easily facilitated by using some standard numerical packages. A numerical example is presented to demonstrate the validity of the proposed learning law.

Original languageEnglish
Article number100201
JournalChinese Physics B
Volume19
Issue number10
DOIs
Publication statusPublished - 2010 Oct 1
Externally publishedYes

Fingerprint

learning
disturbances
norms
formulations

Keywords

  • ℒ ℒ learning law
  • Dynamic neural networks
  • Linear matrix inequality
  • Lyapunov stability theory

ASJC Scopus subject areas

  • Physics and Astronomy(all)

Cite this

2 learning of dynamic neural networks. / Ahn, Choon Ki.

In: Chinese Physics B, Vol. 19, No. 10, 100201, 01.10.2010.

Research output: Contribution to journalArticle

@article{48d2643c52224ca9886ca2de453998d6,
title = "ℒ2 ℒ∞ learning of dynamic neural networks",
abstract = "This paper proposes an ℒ2 ℒ∞ learning law as a new learning method for dynamic neural networks with external disturbance. Based on linear matrix inequality (LMI) formulation, the ℒ2 ℒ∞ learning law is presented to not only guarantee asymptotical stability of dynamic neural networks but also reduce the effect of external disturbance to an ℒ2 ℒ∞ induced norm constraint. It is shown that the design of the ℒ2 ℒ∞ learning law for such neural networks can be achieved by solving LMIs, which can be easily facilitated by using some standard numerical packages. A numerical example is presented to demonstrate the validity of the proposed learning law.",
keywords = "ℒ ℒ learning law, Dynamic neural networks, Linear matrix inequality, Lyapunov stability theory",
author = "Ahn, {Choon Ki}",
year = "2010",
month = "10",
day = "1",
doi = "10.1088/1674-1056/19/10/100201",
language = "English",
volume = "19",
journal = "Chinese Physics B",
issn = "1674-1056",
publisher = "IOP Publishing Ltd.",
number = "10",

}

TY - JOUR

T1 - ℒ2 ℒ∞ learning of dynamic neural networks

AU - Ahn, Choon Ki

PY - 2010/10/1

Y1 - 2010/10/1

N2 - This paper proposes an ℒ2 ℒ∞ learning law as a new learning method for dynamic neural networks with external disturbance. Based on linear matrix inequality (LMI) formulation, the ℒ2 ℒ∞ learning law is presented to not only guarantee asymptotical stability of dynamic neural networks but also reduce the effect of external disturbance to an ℒ2 ℒ∞ induced norm constraint. It is shown that the design of the ℒ2 ℒ∞ learning law for such neural networks can be achieved by solving LMIs, which can be easily facilitated by using some standard numerical packages. A numerical example is presented to demonstrate the validity of the proposed learning law.

AB - This paper proposes an ℒ2 ℒ∞ learning law as a new learning method for dynamic neural networks with external disturbance. Based on linear matrix inequality (LMI) formulation, the ℒ2 ℒ∞ learning law is presented to not only guarantee asymptotical stability of dynamic neural networks but also reduce the effect of external disturbance to an ℒ2 ℒ∞ induced norm constraint. It is shown that the design of the ℒ2 ℒ∞ learning law for such neural networks can be achieved by solving LMIs, which can be easily facilitated by using some standard numerical packages. A numerical example is presented to demonstrate the validity of the proposed learning law.

KW - ℒ ℒ learning law

KW - Dynamic neural networks

KW - Linear matrix inequality

KW - Lyapunov stability theory

UR - http://www.scopus.com/inward/record.url?scp=78649338805&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78649338805&partnerID=8YFLogxK

U2 - 10.1088/1674-1056/19/10/100201

DO - 10.1088/1674-1056/19/10/100201

M3 - Article

AN - SCOPUS:78649338805

VL - 19

JO - Chinese Physics B

JF - Chinese Physics B

SN - 1674-1056

IS - 10

M1 - 100201

ER -