Smart Redundancy Schemes for ANNs against Fault Attacks

Troya Çağıl  Köylü; Said Hamdioui; Mottaqiallah Taouil

doi:10.1109/ETS54262.2022.9810380

Smart Redundancy Schemes for ANNs against Fault Attacks

Troya Çağıl Köylü, Said Hamdioui, Mottaqiallah Taouil

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

1 Citation (Scopus)

45 Downloads (Pure)

Abstract

Artificial neural networks (ANNs) are used to accomplish a variety of tasks, including safety critical ones. Hence, it is important to protect them against faults that can influence decisions during operation. In this paper, we propose smart and low-cost redundancy schemes that protect the most vulnerable ANN parts against fault attacks. Experimental results show that the two proposed smart schemes perform similarly to dual modular redundancy (DMR) at a much lower cost, generally improve on the state of the art, and reach protection levels in the range of 93% to 99%.

Original language	English
Title of host publication	Proceedings of the 2022 IEEE European Test Symposium (ETS)
Place of Publication	Danvers
Publisher	IEEE
Pages	1-2
Number of pages	2
ISBN (Electronic)	978-1-6654-6706-3
ISBN (Print)	978-1-6654-6707-0
DOIs	https://doi.org/10.1109/ETS54262.2022.9810380
Publication status	Published - 2022
Event	2022 IEEE European Test Symposium (ETS) - Barcelona, Spain Duration: 23 May 2022 → 27 May 2022

Conference

Conference	2022 IEEE European Test Symposium (ETS)
Country/Territory	Spain
City	Barcelona
Period	23/05/22 → 27/05/22

Bibliographical note

Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care
Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Keywords

artificial neural network
redundancy
fault injection
countermeasure
machine learning

Access to Document

10.1109/ETS54262.2022.9810380

Smart_Redundancy_Schemes_for_ANNs_Against_Fault_AttacksFinal published version, 1.05 MB

Cite this

@inproceedings{0d66fe27a2354775857ac49e9c59976a,

title = "Smart Redundancy Schemes for ANNs against Fault Attacks",

abstract = "Artificial neural networks (ANNs) are used to accomplish a variety of tasks, including safety critical ones. Hence, it is important to protect them against faults that can influence decisions during operation. In this paper, we propose smart and low-cost redundancy schemes that protect the most vulnerable ANN parts against fault attacks. Experimental results show that the two proposed smart schemes perform similarly to dual modular redundancy (DMR) at a much lower cost, generally improve on the state of the art, and reach protection levels in the range of 93% to 99%.",

keywords = "artificial neural network, redundancy, fault injection, countermeasure, machine learning",

author = "K{\"o}yl{\"u}, {Troya {\c C}ağıl} and Said Hamdioui and Mottaqiallah Taouil",

note = "Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.; 2022 IEEE European Test Symposium (ETS) ; Conference date: 23-05-2022 Through 27-05-2022",

year = "2022",

doi = "10.1109/ETS54262.2022.9810380",

language = "English",

isbn = "978-1-6654-6707-0",

pages = "1--2",

booktitle = "Proceedings of the 2022 IEEE European Test Symposium (ETS)",

publisher = "IEEE",

address = "United States",

}

TY - GEN

T1 - Smart Redundancy Schemes for ANNs against Fault Attacks

AU - Köylü, Troya Çağıl

AU - Hamdioui, Said

AU - Taouil, Mottaqiallah

N1 - Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

PY - 2022

Y1 - 2022

N2 - Artificial neural networks (ANNs) are used to accomplish a variety of tasks, including safety critical ones. Hence, it is important to protect them against faults that can influence decisions during operation. In this paper, we propose smart and low-cost redundancy schemes that protect the most vulnerable ANN parts against fault attacks. Experimental results show that the two proposed smart schemes perform similarly to dual modular redundancy (DMR) at a much lower cost, generally improve on the state of the art, and reach protection levels in the range of 93% to 99%.

AB - Artificial neural networks (ANNs) are used to accomplish a variety of tasks, including safety critical ones. Hence, it is important to protect them against faults that can influence decisions during operation. In this paper, we propose smart and low-cost redundancy schemes that protect the most vulnerable ANN parts against fault attacks. Experimental results show that the two proposed smart schemes perform similarly to dual modular redundancy (DMR) at a much lower cost, generally improve on the state of the art, and reach protection levels in the range of 93% to 99%.

KW - artificial neural network

KW - redundancy

KW - fault injection

KW - countermeasure

KW - machine learning

UR - http://www.scopus.com/inward/record.url?scp=85134220294&partnerID=8YFLogxK

U2 - 10.1109/ETS54262.2022.9810380

DO - 10.1109/ETS54262.2022.9810380

M3 - Conference contribution

SN - 978-1-6654-6707-0

SP - 1

EP - 2

BT - Proceedings of the 2022 IEEE European Test Symposium (ETS)

PB - IEEE

CY - Danvers

T2 - 2022 IEEE European Test Symposium (ETS)

Y2 - 23 May 2022 through 27 May 2022

ER -

Smart Redundancy Schemes for ANNs against Fault Attacks

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this