Back to Search Start Over

Emulating the Effects of Radiation-Induced Soft-Errors for the Reliability Assessment of Neural Networks

Authors :
Luigi Dilillo
Carlo Cazzaniga
Annachiara Ruospo
Alberto Bosio
Daniel Soderstrom
Maria Kastriotou
Ernesto Sanchez
Lucas Matanaluza
Université de Montpellier (UM)
Politecnico di Torino = Polytechnic of Turin (Polito)
University of Jyväskylä (JYU)
​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​ISIS Neutron and Muon Source (ISIS)
STFC Rutherford Appleton Laboratory (RAL)
Science and Technology Facilities Council (STFC)-Science and Technology Facilities Council (STFC)
Science and Technology Facilities Council (STFC)
Institut des Nanotechnologies de Lyon (INL)
École Centrale de Lyon (ECL)
Université de Lyon-Université de Lyon-Université Claude Bernard Lyon 1 (UCBL)
Université de Lyon-École supérieure de Chimie Physique Electronique de Lyon (CPE)-Institut National des Sciences Appliquées de Lyon (INSA Lyon)
Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS)
TEST (TEST)
Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM)
Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)
Source :
IEEE Transactions on Emerging Topics in Computing, IEEE Transactions on Emerging Topics in Computing, Institute of Electrical and Electronics Engineers, In press, pp.1-1. ⟨10.1109/TETC.2021.3116999⟩
Publication Year :
2022
Publisher :
Institute of Electrical and Electronics Engineers (IEEE), 2022.

Abstract

International audience; Convolutional Neural Networks (CNNs) are currently one of the most widely used predictive models in machine learning. Recent studies have demonstrated that hardware faults induced by radiation fields, including cosmic rays, may significantly impact the CNN inference leading to wrong predictions. Therefore, ensuring the reliability of CNNs is crucial, especially for safety-critical systems. In the literature, several works propose reliability assessments of CNNs mainly based on statistically injected faults. This work presents a software emulator capable of injecting real faults retrieved from radiation tests. Specifically, from the device characterisation of a DRAM memory, we extracted event rates and fault models. The software emulator can reproduce their incidence and access their effect on CNN applications with a reliability assessment precision close to the physical one. Radiation-based physical injections and emulator-based injections are performed on three CNNs (LeNet-5) exploiting different data representations. Their outcomes are compared, and the software results evidence that the emulator is able to reproduce the faulty behaviours observed during the radiation tests for the targeted CNNs. This approach leads to a more concise use of radiation experiments since the extracted fault models can be reused to explore different scenarios (e.g., impact on a different application).

Details

Language :
English
ISSN :
21686750
Database :
OpenAIRE
Journal :
IEEE Transactions on Emerging Topics in Computing, IEEE Transactions on Emerging Topics in Computing, Institute of Electrical and Electronics Engineers, In press, pp.1-1. ⟨10.1109/TETC.2021.3116999⟩
Accession number :
edsair.doi.dedup.....92ce93a1d8750deeb0a9765f433053e1
Full Text :
https://doi.org/10.1109/TETC.2021.3116999⟩