Implementing Convolutional Neural Networks Using Hartley Stochastic Computing With Adaptive Rate Feature Map Compression

Implementing Convolutional Neural Networks Using Hartley Stochastic Computing With Adaptive Rate Feature Map Compression

Energy consumption and the latency of convolutional neural networks (CNNs) are two important factors that limit their applications specifically for embedded devices. Fourier-based frequency domain (FD) convolution is a promising low-cost alternative to conventional implementations in the spatial dom...

Full description

Bibliographic Details
Main Authors: S. H. Mozafari, J. J. Clark, W. J. Gross, B. H. Meyer
Format: Article
Language:English
Published: IEEE 2021-01-01
Series:IEEE Open Journal of Circuits and Systems
Subjects:
Deep neural networks
frequency domain transformation
hardware implementation
energy optimization
latency improvement
FPGA
Online Access:https://ieeexplore.ieee.org/document/9645050/

Internet

https://ieeexplore.ieee.org/document/9645050/

Similar Items