Robust stability of recurrent neural networks with ISS learning algorithm

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

In this paper, an input-to-state stability (ISS) approach is used to derive a new robust weight learning algorithm for dynamic neural networks with external disturbance. Based on linear matrix inequality (LMI) formulation, the ISS learning algorithm is presented to not only guarantee exponential stability but also reduce the effect of an external disturbance. It is shown that the design of the ISS learning algorithm can be achieved by solving LMI, which can be easily facilitated by using some standard numerical packages. A numerical example is presented to demonstrate the validity of the proposed learning algorithm.

Original languageEnglish
Pages (from-to)413-419
Number of pages7
JournalNonlinear Dynamics
Volume65
Issue number4
DOIs
Publication statusPublished - 2011 Sep 1
Externally publishedYes

Fingerprint

Recurrent neural networks
Recurrent Neural Networks
Robust Stability
Learning algorithms
Learning Algorithm
Linear matrix inequalities
Matrix Inequality
Linear Inequalities
Disturbance
Dynamic Neural Networks
Exponential Stability
Asymptotic stability
Neural networks
Numerical Examples
Robust stability
Formulation
Demonstrate

Keywords

  • Dynamic neural networks
  • Input-to-state stability (ISS) approach
  • Linear matrix inequality (LMI)
  • Weight learning algorithm

ASJC Scopus subject areas

  • Applied Mathematics
  • Mechanical Engineering
  • Aerospace Engineering
  • Ocean Engineering
  • Electrical and Electronic Engineering
  • Control and Systems Engineering

Cite this

Robust stability of recurrent neural networks with ISS learning algorithm. / Ahn, Choon Ki.

In: Nonlinear Dynamics, Vol. 65, No. 4, 01.09.2011, p. 413-419.

Research output: Contribution to journalArticle

@article{a96b685c5fe1484abdaeab32593b5a9b,
title = "Robust stability of recurrent neural networks with ISS learning algorithm",
abstract = "In this paper, an input-to-state stability (ISS) approach is used to derive a new robust weight learning algorithm for dynamic neural networks with external disturbance. Based on linear matrix inequality (LMI) formulation, the ISS learning algorithm is presented to not only guarantee exponential stability but also reduce the effect of an external disturbance. It is shown that the design of the ISS learning algorithm can be achieved by solving LMI, which can be easily facilitated by using some standard numerical packages. A numerical example is presented to demonstrate the validity of the proposed learning algorithm.",
keywords = "Dynamic neural networks, Input-to-state stability (ISS) approach, Linear matrix inequality (LMI), Weight learning algorithm",
author = "Ahn, {Choon Ki}",
year = "2011",
month = "9",
day = "1",
doi = "10.1007/s11071-010-9901-5",
language = "English",
volume = "65",
pages = "413--419",
journal = "Nonlinear Dynamics",
issn = "0924-090X",
publisher = "Springer Netherlands",
number = "4",

}

TY - JOUR

T1 - Robust stability of recurrent neural networks with ISS learning algorithm

AU - Ahn, Choon Ki

PY - 2011/9/1

Y1 - 2011/9/1

N2 - In this paper, an input-to-state stability (ISS) approach is used to derive a new robust weight learning algorithm for dynamic neural networks with external disturbance. Based on linear matrix inequality (LMI) formulation, the ISS learning algorithm is presented to not only guarantee exponential stability but also reduce the effect of an external disturbance. It is shown that the design of the ISS learning algorithm can be achieved by solving LMI, which can be easily facilitated by using some standard numerical packages. A numerical example is presented to demonstrate the validity of the proposed learning algorithm.

AB - In this paper, an input-to-state stability (ISS) approach is used to derive a new robust weight learning algorithm for dynamic neural networks with external disturbance. Based on linear matrix inequality (LMI) formulation, the ISS learning algorithm is presented to not only guarantee exponential stability but also reduce the effect of an external disturbance. It is shown that the design of the ISS learning algorithm can be achieved by solving LMI, which can be easily facilitated by using some standard numerical packages. A numerical example is presented to demonstrate the validity of the proposed learning algorithm.

KW - Dynamic neural networks

KW - Input-to-state stability (ISS) approach

KW - Linear matrix inequality (LMI)

KW - Weight learning algorithm

UR - http://www.scopus.com/inward/record.url?scp=84863334042&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84863334042&partnerID=8YFLogxK

U2 - 10.1007/s11071-010-9901-5

DO - 10.1007/s11071-010-9901-5

M3 - Article

AN - SCOPUS:84863334042

VL - 65

SP - 413

EP - 419

JO - Nonlinear Dynamics

JF - Nonlinear Dynamics

SN - 0924-090X

IS - 4

ER -