scholar.google.com › citations
According to our experiments, winograd convolution can be utilized to reduce fault-tolerant design overhead by 27.49% or energy consumption by 7.19% without any accuracy loss compared to that without being aware of the fault tolerance.
Feb 17, 2022
Aug 23, 2022 · According to our experiments, winograd convolution can be utilized to reduce fault-tolerant design overhead by 27.49% or energy consumption by ...
Feb 17, 2022 · Abstract:Winograd convolution is originally proposed to reduce the computing overhead by converting multiplication in neural network (NN) ...
According to these experiments, winograd convolution can be utilized to reduce fault-tolerant design overhead or energy consumption without any accuracy ...
... Winograd convolution algorithm [21] is a kind of minimum filtering algorithm that can reduce the times of multiplication by a series of transformations on ...
An Efficient Fault-Tolerant Winograd Convolution for ... - IEEE Xplore
ieeexplore.ieee.org › document
Winograd Convolution has been proved to possess both potentialities on acceleration and fault tolerance. However the state-of-the-art ternary modular redundancy ...
Missing: perspective | Show results with:perspective
Sep 1, 2023 · In this work, we observe the great potential of winograd convolution (WG-Conv) in improving neural network (NN) fault tolerance.
Missing: perspective | Show results with:perspective
People also ask
What is Winograd convolution?
What is fault tolerance of artificial neural network?
Winograd Convolution: A Perspective from Fault Tolerance ... Winograd convolution is originally proposed to reduce the computing overhead by converting ...
Jan 31, 2024 · According to our experiments, winograd convolution can reduce the fault-tolerant design overhead by 55.77\% on average without any accuracy loss ...
According to our experiments, winograd convolution can reduce the fault-tolerant design overhead by 55.77\% on average without any accuracy loss compared to ...