A survey on modeling and improving reliability of DNN algorithms and accelerators

• Published in: Journal of systems architecture Vol. 104; p. 101689
• Main Author: Mittal, Sparsh
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.03.2020
• Subjects: Deep learning; Deep neural networks; Fault-injection; Permanent fault; Review; Transient fault; Deep learning; Deep neural networks; Review; Fault-injection; Permanent fault; Transient fault

As DNNs become increasingly common in mission-critical applications, ensuring their reliable operation has become crucial. Conventional resilience techniques fail to account for the unique characteristics of DNN algorithms/accelerators, and hence, they are infeasible or ineffective. In this paper, w...