On Neural Networks Redundancy and Diversity for Their Use in Safety-Critical Systems

被引：3

作者：

Brando, Axel ^{[1
]}

Serra, Isabel ^{[2
]}

Mezzetti, Enrico ^{[1
]}

Cazorla, Francisco J. ^{[3
]}

Perez-Cerrolaza, Jon ^{[4
]}

Abella, Jaume ^{[3
]}

机构：

[1] Barcelona Supercomp Ctr, Barcelona 08034, Spain

[2] Barcelona Supercomp Ctr, CAOS Grp, Barcelona 08034, Spain

[3] Barcelona Supercomp Ctr, Grp Interact Operating Syst & Comp Architecture C, Barcelona 08034, Spain

[4] Ikerlan Technol Res Ctr, Arrasate Mondragon 20500, Spain

来源：

COMPUTER | 2023年 / 56卷 / 05期

基金：

欧洲研究理事会;

关键词：

Deep learning; Redundancy; Neural networks; Safety; Standards;

D O I：

10.1109/MC.2023.3236523

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

An increasing number of critical functionalities integrated in embedded critical systems rely on deep learning (DL) technology. This article summarizes certain key aspects of DL's intrinsic stochastic and training-data-dependent nature that are at odds with current domain-specific functional safety standards. We exemplify how redundancy and diversity of neural networks can help to reconcile DL technology and functional safety requirements.

引用

页码：41 / 50

页数：10

共 50 条

[1] Utilising Redundancy to Enhance Security of Safety-Critical Systems
Troubitsyna, Elena
[J]. COMPUTER SAFETY, RELIABILITY, AND SECURITY, SAFECOMP 2023 WORKSHOPS, 2023, 14182 : 188 - 196
[2] Verification and validation of neural networks for safety-critical applications
Hull, J
Ward, D
Zakrzewski, RR
[J]. PROCEEDINGS OF THE 2002 AMERICAN CONTROL CONFERENCE, VOLS 1-6, 2002, 1-6 : 4789 - 4794
[3] Challenges in Using Neural Networks in Safety-Critical Applications
Forsberg, Hakan
Linden, Joakim
Hjorth, Johan
Manefjord, Torbjom
Daneshtalab, Masoud
[J]. 2020 AIAA/IEEE 39TH DIGITAL AVIONICS SYSTEMS CONFERENCE (DASC) PROCEEDINGS, 2020,
[4] Experimental Evaluation of Kalman Predictive Redundancy System for Safety-Critical Systems
Kim, Man Ho
Park, Jee Hun
Lee, Suk
Lee, Kyung Chang
[J]. ISIE: 2009 IEEE INTERNATIONAL SYMPOSIUM ON INDUSTRIAL ELECTRONICS, 2009, : 1293 - +
[5] Kalman Predictive Redundancy System for Fault Tolerance of Safety-Critical Systems
Kim, Man Ho
Lee, Suk
Lee, Kyung Chang
[J]. IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, 2010, 6 (01) : 46 - 53
[6] THE USE OF MULTICHIP MODULES IN SAFETY-CRITICAL SYSTEMS
ROUGHTON, M
HINDE, G
[J]. ELECTRONIC ENGINEERING, 1994, 66 (809): : 35 - &
[7] SAFETY-CRITICAL SYSTEMS
MCDERMID, JA
THEWLIS, DJ
[J]. SOFTWARE ENGINEERING JOURNAL, 1991, 6 (02): : 35 - 35
[8] SAFETY-CRITICAL SYSTEMS
RUDALL, BH
[J]. ROBOTICA, 1990, 8 : 184 - 184
[9] Neural Networks for Safety-Critical Applications - Challenges, Fxperiments and Perspectives
Cheng, Chih-Hong
Diehl, Frederik
Hinz, Gereon
Hamza, Yassine
Nuehrenberg, Georg
Rickert, Markus
Ruess, Harald
Truong-Le, Michael
[J]. PROCEEDINGS OF THE 2018 DESIGN, AUTOMATION & TEST IN EUROPE CONFERENCE & EXHIBITION (DATE), 2018, : 1005 - 1006
[10] THE CHOICE OF COMPUTER LANGUAGES FOR USE IN SAFETY-CRITICAL SYSTEMS
CULLYER, WJ
GOODENOUGH, SJ
WICHMANN, BA
[J]. SOFTWARE ENGINEERING JOURNAL, 1991, 6 (02): : 51 - 58

← 1 2 3 4 5 →