Skip to Main content Skip to Navigation
Conference papers

A Reliability Study on CNNs for Critical Embedded Systems

Abstract : Deep learning systems such as Convolutional Neural Networks (CNNs) have shown remarkable efficiency in dealing with a variety of complex real life problems. To accelerate the execution of these heavy algorithms, a plethora of software implementations and hardware accelerators have been proposed. In a context of shrinking devices dimensions, reliability issues of CNN-hosting systems are under-explored. In this paper, we experimentally evaluate the inherent fault tolerance of CNNs by injecting errors within network modules, namely processing elements and memories. Our experiments demonstrate a non uniform sensitivity between different parts of the system. While CNNs are relatively resilient to errors occurring in processing elements, transient faults hitting memories lead to catastrophic degradation of accuracy.
Document type :
Conference papers
Complete list of metadata

https://hal-uphf.archives-ouvertes.fr/hal-03430208
Contributor : Kathleen Torck Connect in order to contact the contributor
Submitted on : Tuesday, November 16, 2021 - 10:14:54 AM
Last modification on : Wednesday, November 17, 2021 - 4:01:30 AM

Identifiers

Collections

Citation

Mohamed Ayoub Neggaz, Pablo Ribalta, Ihsen Alouani, Smail Niar. A Reliability Study on CNNs for Critical Embedded Systems. 2018 IEEE 36th International Conference on Computer Design (ICCD), Oct 2018, Orlando, United States. pp.476-479, ⟨10.1109/ICCD.2018.00077⟩. ⟨hal-03430208⟩

Share

Metrics

Record views

8