A Survey on Evolutionary Neural Architecture Search

被引：267

作者：

Liu, Yuqiao ^{[1
]}

Sun, Yanan ^{[1
,2
]}

Xue, Bing ^{[3
]}

Zhang, Mengjie ^{[3
]}

Yen, Gary G. ^{[4
]}

Tan, Kay Chen ^{[5
]}

机构：

[1] Sichuan Univ, Coll Comp Sci, Chengdu 610065, Peoples R China

[2] Nanjing Univ, State Key Lab Novel Software Technol, Nanjing 210023, Peoples R China

[3] Victoria Univ Wellington, Sch Engn & Comp Sci, Wellington 6012, New Zealand

[4] Oklahoma State Univ, Sch Elect & Comp Engn, Stillwater, OK 74078 USA

[5] Hong Kong Polytech Univ, Dept Comp, Hong Kong, Peoples R China

来源：

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS | 2023年 / 34卷 / 02期

基金：

中国国家自然科学基金;

关键词：

Computer architecture; Optimization; Convolutional neural networks; Search problems; Neural networks; Deep learning; Statistics; evolutionary computation (EC); evolutionary neural architecture search (NAS); image classification; PARTICLE SWARM OPTIMIZATION; SHORT-TERM-MEMORY; GENETIC ALGORITHM; NETWORKS; RECOGNITION; COLONY;

D O I：

10.1109/TNNLS.2021.3100554

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Deep neural networks (DNNs) have achieved great success in many applications. The architectures of DNNs play a crucial role in their performance, which is usually manually designed with rich expertise. However, such a design process is labor-intensive because of the trial-and-error process and also not easy to realize due to the rare expertise in practice. Neural architecture search (NAS) is a type of technology that can design the architectures automatically. Among different methods to realize NAS, the evolutionary computation (EC) methods have recently gained much attention and success. Unfortunately, there has not yet been a comprehensive summary of the EC-based NAS algorithms. This article reviews over 200 articles of most recent EC-based NAS methods in light of the core components, to systematically discuss their design principles and justifications on the design. Furthermore, current challenges and issues are also discussed to identify future research in this emerging field.

引用

页码：550 / 570

页数：21

共 228 条

[31]

Chen K., 2020, ARXIV200212704

[32] Equivalence among Stochastic Logic Circuits and its Application [J].

Chen, Te-Hsuan ;

Hayes, John P. .

2015 52ND ACM/EDAC/IEEE DESIGN AUTOMATION CONFERENCE (DAC), 2015,

[33]

Chen Y., 2019, P INT C BIOINSP COMP, P497

[34]

Chen Y., 2018, ARXIV181109426

[35]

Chen Y., 2018, ARXIV180800193

[36]

Chen ZF, 2019, IEEE SYS MAN CYBERN, P3895, DOI [10.1109/SMC.2019.8914267, 10.1109/smc.2019.8914267]

[37] Searching Toward Pareto-Optimal Device-Aware Neural Architectures [J].

Cheng, An-Chieh ;

Dong, Jin-Dong ;

Hsu, Chi-Hung ;

Chang, Shu-Huan ;

Sun, Min ;

Chang, Shih-Chieh ;

Pan, Jia-Yu ;

Chen, Yu-Ting ;

Wei, Wei ;

Juan, Da-Cheng .

2018 IEEE/ACM INTERNATIONAL CONFERENCE ON COMPUTER-AIDED DESIGN (ICCAD) DIGEST OF TECHNICAL PAPERS, 2018,

[38] Facial Expression Recognition in JAFFE Dataset Based on Gaussian Process Classification [J].

Cheng, Fei ;

Yu, Jiangsheng ;

Xiong, Huilin .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 2010, 21 (10) :1685-1690

[39]

Cheung B., 2011, Proceedings of the 2011 Tenth International Conference on Machine Learning and Applications (ICMLA 2011), P293, DOI 10.1109/ICMLA.2011.73

[40]

Chrabaszcz Patryk, 2017, DOWNSAMPLED VARIANT

← 1 2 3 4 5 6 7 8 9 10 →