Information-gap robustness of a neural network regression model

S.G. Pierce, K. Worden, G. Manson

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

Abstract

As a result of their black-box nature, neural networks resist traditional methods of certification and therefore cannot be used in safety critical applications. This situation is undesirable as neural networks can provide an effective solution to many engineering problems. The object of the current paper is to explore the possibility of quantifying and qualifying the reliability of neural networks by a means outside the traditional framework. The approach used here will follow Ben-Haim’s information-gap theory of uncertainty. This is a non-probabilistic approach which may lend itself well to certification of black-box systems. The approach is demonstrated here on a neural network regression model of the process of pre-sliding friction between solids.
LanguageEnglish
Title of host publication22nd IMAC Conference and Exposition 2004 (IMAC XXII): A Conference and Exposition on Structural Dynamics
Pages1068-1076
Number of pages9
Publication statusPublished - 2004
Event22nd IMAC Conference and Exposition 2004 (IMAC XXII) - Dearborn, Michigan, United States
Duration: 26 Jan 200429 Jan 2004

Conference

Conference22nd IMAC Conference and Exposition 2004 (IMAC XXII)
CountryUnited States
CityDearborn, Michigan
Period26/01/0429/01/04

Fingerprint

Neural networks
Friction
Uncertainty

Keywords

  • information-gap
  • robustness
  • neural network
  • regression model

Cite this

Pierce, S. G., Worden, K., & Manson, G. (2004). Information-gap robustness of a neural network regression model. In 22nd IMAC Conference and Exposition 2004 (IMAC XXII): A Conference and Exposition on Structural Dynamics (pp. 1068-1076)
Pierce, S.G. ; Worden, K. ; Manson, G. / Information-gap robustness of a neural network regression model. 22nd IMAC Conference and Exposition 2004 (IMAC XXII): A Conference and Exposition on Structural Dynamics. 2004. pp. 1068-1076
@inproceedings{516ae72e83e54971ad148b520678b501,
title = "Information-gap robustness of a neural network regression model",
abstract = "As a result of their black-box nature, neural networks resist traditional methods of certification and therefore cannot be used in safety critical applications. This situation is undesirable as neural networks can provide an effective solution to many engineering problems. The object of the current paper is to explore the possibility of quantifying and qualifying the reliability of neural networks by a means outside the traditional framework. The approach used here will follow Ben-Haim’s information-gap theory of uncertainty. This is a non-probabilistic approach which may lend itself well to certification of black-box systems. The approach is demonstrated here on a neural network regression model of the process of pre-sliding friction between solids.",
keywords = "information-gap, robustness, neural network, regression model",
author = "S.G. Pierce and K. Worden and G. Manson",
year = "2004",
language = "English",
isbn = "9781604238020",
pages = "1068--1076",
booktitle = "22nd IMAC Conference and Exposition 2004 (IMAC XXII): A Conference and Exposition on Structural Dynamics",

}

Pierce, SG, Worden, K & Manson, G 2004, Information-gap robustness of a neural network regression model. in 22nd IMAC Conference and Exposition 2004 (IMAC XXII): A Conference and Exposition on Structural Dynamics. pp. 1068-1076, 22nd IMAC Conference and Exposition 2004 (IMAC XXII) , Dearborn, Michigan, United States, 26/01/04.

Information-gap robustness of a neural network regression model. / Pierce, S.G.; Worden, K.; Manson, G.

22nd IMAC Conference and Exposition 2004 (IMAC XXII): A Conference and Exposition on Structural Dynamics. 2004. p. 1068-1076.

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

TY - GEN

T1 - Information-gap robustness of a neural network regression model

AU - Pierce, S.G.

AU - Worden, K.

AU - Manson, G.

PY - 2004

Y1 - 2004

N2 - As a result of their black-box nature, neural networks resist traditional methods of certification and therefore cannot be used in safety critical applications. This situation is undesirable as neural networks can provide an effective solution to many engineering problems. The object of the current paper is to explore the possibility of quantifying and qualifying the reliability of neural networks by a means outside the traditional framework. The approach used here will follow Ben-Haim’s information-gap theory of uncertainty. This is a non-probabilistic approach which may lend itself well to certification of black-box systems. The approach is demonstrated here on a neural network regression model of the process of pre-sliding friction between solids.

AB - As a result of their black-box nature, neural networks resist traditional methods of certification and therefore cannot be used in safety critical applications. This situation is undesirable as neural networks can provide an effective solution to many engineering problems. The object of the current paper is to explore the possibility of quantifying and qualifying the reliability of neural networks by a means outside the traditional framework. The approach used here will follow Ben-Haim’s information-gap theory of uncertainty. This is a non-probabilistic approach which may lend itself well to certification of black-box systems. The approach is demonstrated here on a neural network regression model of the process of pre-sliding friction between solids.

KW - information-gap

KW - robustness

KW - neural network

KW - regression model

UR - http://www.sem.org/Proceedings/ConferencePapers-Paper.cfm?ConfPapersPaperID=24505

M3 - Conference contribution book

SN - 9781604238020

SP - 1068

EP - 1076

BT - 22nd IMAC Conference and Exposition 2004 (IMAC XXII): A Conference and Exposition on Structural Dynamics

ER -

Pierce SG, Worden K, Manson G. Information-gap robustness of a neural network regression model. In 22nd IMAC Conference and Exposition 2004 (IMAC XXII): A Conference and Exposition on Structural Dynamics. 2004. p. 1068-1076