Dansk
Reis, Ricardo
(forfatter)
Early Soft Error Reliability Assessment of Convolutional Neural Networks Executing on Resource-Constrained IoT Edge Devices e-bog
656,09 DKK
This book describes an extensive and consistent soft error assessment of convolutional neural network (CNN) models from different domains through more than 14.8 million fault injections, considering different precision bit-width configurations, optimization parameters, and processor models. The authors also evaluate the relative performance, memory utilization, and soft error reliability trade-of…
This book describes an extensive and consistent soft error assessment of convolutional neural network (CNN) models from different domains through more than 14.8 million fault injections, considering different precision bit-width configurations, optimization parameters, and processor models. The authors also evaluate the relative performance, memory utilization, and soft error reliability trade-offs analysis of different CNN models considering a compiler-based technique w.r.t. traditional redundancy approaches.
E-bog
656,09 DKK
Forlag
Springer
Udgivet
01.01.2023
Genrer
Cybernetics and systems theory
Sprog
English
Format
epub
Beskyttelse
LCP
ISBN
9783031185991
This book describes an extensive and consistent soft error assessment of convolutional neural network (CNN) models from different domains through more than 14.8 million fault injections, considering different precision bit-width configurations, optimization parameters, and processor models. The authors also evaluate the relative performance, memory utilization, and soft error reliability trade-offs analysis of different CNN models considering a compiler-based technique w.r.t. traditional redundancy approaches.