Neuron Fault Tolerance in Spiking Neural Networks - Circuits Intégrés Numériques et Analogiques Accéder directement au contenu
Communication Dans Un Congrès Année : 2021

Neuron Fault Tolerance in Spiking Neural Networks

Résumé

The error-resiliency of Artificial Intelligence (AI) hardware accelerators is a major concern, especially when they are deployed in mission-critical and safety-critical applications. In this paper, we propose a neuron fault tolerance strategy for Spiking Neural Networks (SNNs). It is optimized for low area and power overhead by leveraging observations made from a largescale fault injection experiment that pinpoints the critical fault types and locations. We describe the fault modeling approach, the fault injection framework, the results of the fault injection experiment, the fault-tolerance strategy, and the fault-tolerant SNN architecture. The idea is demonstrated on two SNNs that we designed for two SNN-oriented datasets, namely the N-MNIST and IBM's DVS128 gesture datasets.
Fichier principal
Vignette du fichier
1401_pdf_upload.pdf (833.6 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03036630 , version 1 (02-12-2020)

Identifiants

Citer

Theofilos Spyrou, Sarah A El-Sayed, Engin Afacan, Luis A Camuñas-Mesa, Bernabé Linares-Barranco, et al.. Neuron Fault Tolerance in Spiking Neural Networks. 2021 Design, Automation & Test in Europe Conference & Exhibition (DATE), Feb 2021, Grenoble (virtuel), France. pp.743-748, ⟨10.23919/DATE51398.2021.9474081⟩. ⟨hal-03036630⟩
267 Consultations
491 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More