A Systematic Literature Review on Hardware Reliability Assessment Methods for Deep Neural Networks
ACM Computing Surveys · 30.06.2024 · 56 (6) · 1-39
DOI: 10.1145/3638242 ↗Kurz gefasst
Systematische Übersicht in der Zeitschrift ACM Computing Surveys, online erschienen am 22.01.2024. Verfasst von Mohammad Hasan Ahmadilivani und 4 weiteren Autor:innen. Die Arbeit wurde bisher 99-mal zitiert (Crossref) und nennt 201 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Information and Communication Technologies (ICT) programme, Estonian Research Council, Estonian-French science and technology cooperation programme PARROT.
Abstract (Originalsprache)
Artificial Intelligence (AI) and, in particular, Machine Learning (ML), have emerged to be utilized in various applications due to their capability to learn how to solve complex problems. Over the past decade, rapid advances in ML have presented Deep Neural Networks (DNNs) consisting of a large number of neurons and layers. DNN Hardware Accelerators (DHAs) are leveraged to deploy DNNs in the target applications. Safety-critical applications, where hardware faults/errors would result in catastrophic consequences, also benefit from DHAs. Therefore, the reliability of DNNs is an essential subject of research. In recent years, several studies have been published accordingly to assess the reliability of DNNs. In this regard, various reliability assessment methods have been proposed on a variety of platforms and applications. Hence, there is a need to summarize the state-of-the-art to identify the gaps in the study of the reliability of DNNs. In this work, we conduct a Systematic Literature Review (SLR) on the reliability assessment methods of DNNs to collect relevant research works as much as possible, present a categorization of them, and address the open challenges. Through this SLR, three kinds of methods for reliability assessment of DNNs are identified, including Fault Injection (FI), Analytical, and Hybrid methods. Since the majority of works assess the DNN reliability by FI, we characterize different approaches and platforms of the FI method comprehensively. Moreover, Analytical and Hybrid methods are propounded. Thus, different reliability assessment methods for DNNs have been elaborated on their conducted DNN platforms and reliability evaluation metrics. Finally, we highlight the advantages and disadvantages of the identified methods and address the open challenges in the research area. We have concluded that Analytical and Hybrid methods are light-weight yet sufficiently accurate and have the potential to be extended in future research and to be utilized in establishing novel DNN reliability assessment frameworks.
Förderung
- Information and Communication Technologies (ICT) programme
- Estonian Research Council
- Estonian-French science and technology cooperation programme PARROT
- Swedish Innovation Agency VINNOVA project SafeDeep
Zitieren
APA
Ahmadilivani, M. H., Taheri, M., Raik, J., Daneshtalab, M., & Jenihhin, M. (2024). A Systematic Literature Review on Hardware Reliability Assessment Methods for Deep Neural Networks. ACM Computing Surveys, 56(6), 1-39. https://doi.org/10.1145/3638242
Vancouver
Ahmadilivani MH, Taheri M, Raik J, Daneshtalab M, Jenihhin M. A Systematic Literature Review on Hardware Reliability Assessment Methods for Deep Neural Networks. ACM Computing Surveys. 2024;56(6):1-39. doi:10.1145/3638242
BibTeX
@article{Ahmadilivani2024,
author = {Ahmadilivani, Mohammad Hasan and Taheri, Mahdi and Raik, Jaan and Daneshtalab, Masoud and Jenihhin, Maksim},
title = {A Systematic Literature Review on Hardware Reliability Assessment Methods for Deep Neural Networks},
journal = {ACM Computing Surveys},
year = {2024},
volume = {56},
number = {6},
pages = {1--39},
doi = {10.1145/3638242}
}
EUVIMED