Spectral-based convolutional neural network without multiple spatial-frequency domain switchings
Recent researches have shown that spectral representation provides a significant speed-up in the massive computation workload of convolution operations in the inference (feed-forward) algorithm of Convolutional Neural Networks (CNNs). This approach results in reducing the computational complexity of...
Saved in:
Main Authors: | Ayat, Sayed Omid, Hani, Mohamed Khalil, Ab. Rahman, Ab. Al-Hadi, Abdellatef, Hamdan |
---|---|
格式: | Article |
出版: |
Elsevier B. V.
2019
|
主题: | |
在线阅读: | http://eprints.utm.my/id/eprint/87452/ http://dx.doi.org/10.1016/j.neucom.2019.06.094 |
标签: |
添加标签
没有标签, 成为第一个标记此记录!
|
相似书籍
-
A compact spectral model for convolutional neural network
由: Ayat, Sayed Omid, et al.
出版: (2023) -
Stochastic computing correlation utilization in convolutional neural network basic functions
由: Abdellatef, Hamdan, et al.
出版: (2018) -
Accurate and compact convolutional neural network based on stochastic computing
由: Abdellatef, Hamdan, et al.
出版: (2022) -
A low-complexity complex-valued activation function for fast and accurate spectral domain convolutional neural network
由: Rizvi, S. M., et al.
出版: (2021) -
Hardware and software co-simulation platform for convolution or correlation based image processing algorithms
由: Ayat, Sayed Omid
出版: (2014)