A Review of Recent Advances of Binary Neural Networks for Edge Computing

Wenyu Zhao; Teli Ma; Xuan Gong; Baochang Zhang; David Doermann

doi:10.1109/JMASS.2020.3034205

A Review of Recent Advances of Binary Neural Networks for Edge Computing

Wenyu Zhao
, Teli Ma
, Xuan Gong
, Baochang Zhang^*
, David Doermann

^*Corresponding author for this work

School of Artificial Intelligence(Institute of Artificial Intelligence)

Research output: Contribution to journal › Review article › peer-review

23 Scopus citations

Abstract

Edge computing is promising to become one of the next hottest topics in artificial intelligence because it benefits various evolving domains, such as real-time unmanned aerial systems, industrial applications, and the demand for privacy protection. This article reviews the recent advances on binary neural network (BNN) and 1-bit convolutional neural network technologies that are well suitable for front-end, edge-based computing. We introduce and summarize existing work and classify them based on gradient approximation, quantization, architecture, loss functions, optimization method, and binary neural architecture search. We also introduce applications in the areas of computer vision and speech recognition and discuss future applications for edge computing.

Original language	English
Article number	9240984
Pages (from-to)	25-35
Number of pages	11
Journal	IEEE Journal on Miniaturization for Air and Space Systems
Volume	2
Issue number	1
DOIs	https://doi.org/10.1109/JMASS.2020.3034205
State	Published - Mar 2021

Keywords

1-bit convolutional neural network (CNN)
binary neural network (BNN)
edge computing
front-end computing
neural architecture search

Access to Document

10.1109/JMASS.2020.3034205

Cite this

@article{deab73c67305457e8821f339bd51bb20,

title = "A Review of Recent Advances of Binary Neural Networks for Edge Computing",

abstract = "Edge computing is promising to become one of the next hottest topics in artificial intelligence because it benefits various evolving domains, such as real-time unmanned aerial systems, industrial applications, and the demand for privacy protection. This article reviews the recent advances on binary neural network (BNN) and 1-bit convolutional neural network technologies that are well suitable for front-end, edge-based computing. We introduce and summarize existing work and classify them based on gradient approximation, quantization, architecture, loss functions, optimization method, and binary neural architecture search. We also introduce applications in the areas of computer vision and speech recognition and discuss future applications for edge computing.",

keywords = "1-bit convolutional neural network (CNN), binary neural network (BNN), edge computing, front-end computing, neural architecture search",

author = "Wenyu Zhao and Teli Ma and Xuan Gong and Baochang Zhang and David Doermann",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.",

year = "2021",

month = mar,

doi = "10.1109/JMASS.2020.3034205",

language = "英语",

volume = "2",

pages = "25--35",

journal = "IEEE Journal on Miniaturization for Air and Space Systems",

issn = "2576-3164",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "1",

}

TY - JOUR

T1 - A Review of Recent Advances of Binary Neural Networks for Edge Computing

AU - Zhao, Wenyu

AU - Ma, Teli

AU - Gong, Xuan

AU - Zhang, Baochang

AU - Doermann, David

PY - 2021/3

Y1 - 2021/3

N2 - Edge computing is promising to become one of the next hottest topics in artificial intelligence because it benefits various evolving domains, such as real-time unmanned aerial systems, industrial applications, and the demand for privacy protection. This article reviews the recent advances on binary neural network (BNN) and 1-bit convolutional neural network technologies that are well suitable for front-end, edge-based computing. We introduce and summarize existing work and classify them based on gradient approximation, quantization, architecture, loss functions, optimization method, and binary neural architecture search. We also introduce applications in the areas of computer vision and speech recognition and discuss future applications for edge computing.

AB - Edge computing is promising to become one of the next hottest topics in artificial intelligence because it benefits various evolving domains, such as real-time unmanned aerial systems, industrial applications, and the demand for privacy protection. This article reviews the recent advances on binary neural network (BNN) and 1-bit convolutional neural network technologies that are well suitable for front-end, edge-based computing. We introduce and summarize existing work and classify them based on gradient approximation, quantization, architecture, loss functions, optimization method, and binary neural architecture search. We also introduce applications in the areas of computer vision and speech recognition and discuss future applications for edge computing.

KW - 1-bit convolutional neural network (CNN)

KW - binary neural network (BNN)

KW - edge computing

KW - front-end computing

KW - neural architecture search

UR - https://www.scopus.com/pages/publications/85177718345

U2 - 10.1109/JMASS.2020.3034205

DO - 10.1109/JMASS.2020.3034205

M3 - 文献综述

AN - SCOPUS:85177718345

SN - 2576-3164

VL - 2

SP - 25

EP - 35

JO - IEEE Journal on Miniaturization for Air and Space Systems

JF - IEEE Journal on Miniaturization for Air and Space Systems

IS - 1

M1 - 9240984

ER -

A Review of Recent Advances of Binary Neural Networks for Edge Computing

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this