MOZART+: Masking Outputs with Zeros for Improved Architectural Robustness and Testing of DNN Accelerators - Département Systèmes et Circuits Intégrés Numériques Accéder directement au contenu
Article Dans Une Revue IEEE Transactions on Device and Materials Reliability Année : 2023

MOZART+: Masking Outputs with Zeros for Improved Architectural Robustness and Testing of DNN Accelerators

Résumé

Deep Neural Networks (DNNs) are increasingly used in safety critical autonomous systems. We present MOZART+, a DNN accelerator architecture which provides fault detection and fault tolerance. MOZART+ is a systolic architecture based on the Output Stationary (OS) data-flow, as it is a data-flow that inherently limits fault propagation. In addition, MOZART+ achieves fault detection with on-line functional testing of the Processing Elements (PEs). Faulty PEs are swiftly taken off-line with minimal classification impact. We show how to handle the case of layers with a small number of neurons. The implementation of our approach on Squeezenet results in a loss of accuracy of less than 3% in the presence of a single faulty PE, compared to 15-33% without mitigation. The area overhead for the test logic does not exceed 8%. Dropout during training further improves fault tolerance, without a priori knowledge of the faults. We present a detailed fault-injection study on multiple systolic architectures, considering different fault-models and comparing different measures of accuracy.
Fichier principal
Vignette du fichier
mozart_tdmr_21_11_24.pdf (722.92 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03823955 , version 1 (04-06-2024)

Identifiants

Citer

Stephane Burel, Adrian Evans, Lorena Anghel. MOZART+: Masking Outputs with Zeros for Improved Architectural Robustness and Testing of DNN Accelerators. IEEE Transactions on Device and Materials Reliability, 2023, 22 (2), pp.120-128. ⟨10.1109/TDMR.2022.3159089⟩. ⟨hal-03823955⟩
30 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Mastodon Facebook X LinkedIn More