Result: Fault Tolerance in Quantized and Pruned Convolutional Neural Networks

Title:
Fault Tolerance in Quantized and Pruned Convolutional Neural Networks
Authors:
Guillemé, Wilfread, Kritikakou, Angeliki, Helen, Youri, Killian, Cédric, Chillet, Daniel
Contributors:
Architectures matérielles spécialisées pour l’ère post loi-de-Moore (TARAN), Centre Inria de l'Université de Rennes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-ARCHITECTURE (IRISA-D3), Institut de Recherche en Informatique et Systèmes Aléatoires (IRISA), Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Université de Bretagne Sud (UBS)-École normale supérieure - Rennes (ENS Rennes)-Institut National de Recherche en Informatique et en Automatique (Inria)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-IMT Atlantique (IMT Atlantique), Institut Mines-Télécom [Paris] (IMT)-Institut Mines-Télécom [Paris] (IMT)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut Mines-Télécom [Paris] (IMT)-Institut Mines-Télécom [Paris] (IMT)-Institut de Recherche en Informatique et Systèmes Aléatoires (IRISA), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Université de Bretagne Sud (UBS)-École normale supérieure - Rennes (ENS Rennes)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-IMT Atlantique (IMT Atlantique), Institut Mines-Télécom [Paris] (IMT)-Institut Mines-Télécom [Paris] (IMT), DGA Maîtrise de l'information (DGA.MI), Direction générale de l'armement [France] (DGA), Laboratoire Hubert Curien (LabHC), Institut d'Optique Graduate School (IOGS)-Université Jean Monnet - Saint-Étienne (UJM)-Centre National de la Recherche Scientifique (CNRS), Inria, Agence de l'innovation de défense
Source:
IOLTS 2025 - IEEE 31st International Symposium on On-Line Testing and Robust System Design. :1-7
Publisher Information:
CCSD; IEEE, 2025.
Publication Year:
2025
Collection:
collection:UNIV-ST-ETIENNE
collection:IOGS
collection:UNIV-RENNES1
collection:CNRS
collection:INRIA
collection:UNIV-UBS
collection:INSA-RENNES
collection:ENSSAT
collection:INRIA-RENNES
collection:IRISA
collection:IRISA_SET
collection:INRIA_TEST
collection:TESTALAIN1
collection:CENTRALESUPELEC
collection:INRIA2
collection:UR1-HAL
collection:UR1-MATH-STIC
collection:UR1-UFR-ISTIC
collection:TEST-UR-CSS
collection:UNIV-RENNES
collection:INRIA-RENGRE
collection:UDL
collection:UR1-MATH-NUM
collection:LABORATOIRE-HUBERT-CURIEN
Subject Terms:
TMR., SEU, Mitigation, Fault Tolerance, CNN, [INFO]Computer Science [cs]
Subject Geographic:
Ischia, Italy
Original Identifier:
HAL: hal-05313661
Document Type:
Conference conferenceObject<br />Conference papers
Language:
English
Relation:
info:eu-repo/semantics/altIdentifier/doi/10.1109/IOLTS65288.2025.11117099
DOI:
10.1109/IOLTS65288.2025.11117099
Access URL:
https://inria.hal.science/hal-05313661
https://inria.hal.science/hal-05313661v1/document
https://inria.hal.science/hal-05313661v1/file/IOLTS25_CameraReady.pdf
Rights:
info:eu-repo/semantics/OpenAccess
URL: http://creativecommons.org/licenses/by/
Accession Number:
edshal.hal.05313661v1
Database:
HAL

Further Information

Convolutional Neural Networks (CNN), particularly those used in critical applications, such as autonomous driving, medical systems, and aerospace, require high reliability. While these algorithms exhibit inherent resilience, they remain susceptible to Single-Event Effects (SEE) occurring at the hardware and impacting the model execution. These effects, usually induced by interactions with radiation particles, can lead to errors in electronic components, potentially causing incorrect inferences and increasing the risk of mispredictions. Meanwhile, quantization and pruning are widely employed to reduce the hardware footprint of CNN models, facilitating their deployment on embedded systems. Even when the models are reduced, CNN remain too large for an exhaustive fault injection campaign to assess their resilience. To address these challenges, we propose SFI4NN, a Statistical Fault Injection (SFI) framework specifically designed to evaluate the fault sensitivity of fixed-point quantized and pruned CNN architectures. Furthermore, we analyze the model resilience as a function of the pruning rate, showing that CNN sensitivity increases as pruning becomes more aggressive. The obtained results enable the development of hardware hardening strategies with reduced costs that are tailored to the reliability requirements of targeted applications. Experimental results demonstrate a 96\% improvement in resilience, with minimal hardware overhead compared to conventional hardening techniques such as triplication.