EEEA-Net: An Early Exit Evolutionary Neural Architecture Search

The goals of this research were to search for Convolutional Neural Network (CNN) architectures, suitable for an on-device processor with limited computing resources, performing at substantially lower Network Architecture Search (NAS) costs. A new algorithm entitled an Early Exit Population Initialis...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Engineering applications of artificial intelligence Jg. 104; S. 104397
Hauptverfasser: Termritthikun, Chakkrit, Jamtsho, Yeshi, Ieamsaard, Jirarat, Muneesawang, Paisarn, Lee, Ivan
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier Ltd 01.09.2021
Schlagworte:
ISSN:0952-1976, 1873-6769
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract The goals of this research were to search for Convolutional Neural Network (CNN) architectures, suitable for an on-device processor with limited computing resources, performing at substantially lower Network Architecture Search (NAS) costs. A new algorithm entitled an Early Exit Population Initialisation (EE-PI) for Evolutionary Algorithm (EA) was developed to achieve both goals. The EE-PI reduces the total number of parameters in the search process by filtering the models with fewer parameters than the maximum threshold. It will look for a new model to replace those models with parameters more than the threshold. Thereby, reducing the number of parameters, memory usage for model storage and processing time while maintaining the same performance or accuracy. The search time was reduced to 0.52 GPU day. This is a huge and significant achievement compared to the NAS of 4 GPU days achieved using NSGA-Net, 3,150 GPU days by the AmoebaNet model, and the 2,000 GPU days by the NASNet model. As well, Early Exit Evolutionary Algorithm networks (EEEA-Nets) yield network architectures with minimal error and computational cost suitable for a given dataset as a class of network algorithms. Using EEEA-Net on CIFAR-10, CIFAR-100, and ImageNet datasets, our experiments showed that EEEA-Net achieved the lowest error rate among state-of-the-art NAS models, with 2.46% for CIFAR-10, 15.02% for CIFAR-100, and 23.8% for ImageNet dataset. Further, we implemented this image recognition architecture for other tasks, such as object detection, semantic segmentation, and keypoint detection tasks, and, in our experiments, EEEA-Net-C2 outperformed MobileNet-V3 on all of these various tasks. (The algorithm code is available at https://github.com/chakkritte/EEEA-Net).
AbstractList The goals of this research were to search for Convolutional Neural Network (CNN) architectures, suitable for an on-device processor with limited computing resources, performing at substantially lower Network Architecture Search (NAS) costs. A new algorithm entitled an Early Exit Population Initialisation (EE-PI) for Evolutionary Algorithm (EA) was developed to achieve both goals. The EE-PI reduces the total number of parameters in the search process by filtering the models with fewer parameters than the maximum threshold. It will look for a new model to replace those models with parameters more than the threshold. Thereby, reducing the number of parameters, memory usage for model storage and processing time while maintaining the same performance or accuracy. The search time was reduced to 0.52 GPU day. This is a huge and significant achievement compared to the NAS of 4 GPU days achieved using NSGA-Net, 3,150 GPU days by the AmoebaNet model, and the 2,000 GPU days by the NASNet model. As well, Early Exit Evolutionary Algorithm networks (EEEA-Nets) yield network architectures with minimal error and computational cost suitable for a given dataset as a class of network algorithms. Using EEEA-Net on CIFAR-10, CIFAR-100, and ImageNet datasets, our experiments showed that EEEA-Net achieved the lowest error rate among state-of-the-art NAS models, with 2.46% for CIFAR-10, 15.02% for CIFAR-100, and 23.8% for ImageNet dataset. Further, we implemented this image recognition architecture for other tasks, such as object detection, semantic segmentation, and keypoint detection tasks, and, in our experiments, EEEA-Net-C2 outperformed MobileNet-V3 on all of these various tasks. (The algorithm code is available at https://github.com/chakkritte/EEEA-Net).
ArticleNumber 104397
Author Muneesawang, Paisarn
Ieamsaard, Jirarat
Lee, Ivan
Termritthikun, Chakkrit
Jamtsho, Yeshi
Author_xml – sequence: 1
  givenname: Chakkrit
  orcidid: 0000-0002-1508-3123
  surname: Termritthikun
  fullname: Termritthikun, Chakkrit
  email: chakkritt60@email.nu.ac.th
  organization: STEM, University of South Australia, Adelaide, SA, 5095, Australia
– sequence: 2
  givenname: Yeshi
  surname: Jamtsho
  fullname: Jamtsho, Yeshi
  email: yjamtsho.cst@rub.edu.bt
  organization: College of Science and Technology, Royal University of Bhutan, Phuentsholing, 21101, Bhutan
– sequence: 3
  givenname: Jirarat
  surname: Ieamsaard
  fullname: Ieamsaard, Jirarat
  email: jirarati@nu.ac.th
  organization: Department of Electrical and Computer Engineering, Faculty of Engineering, Naresuan University, Phitsanulok, 65000, Thailand
– sequence: 4
  givenname: Paisarn
  surname: Muneesawang
  fullname: Muneesawang, Paisarn
  email: paisarnmu@nu.ac.th
  organization: Department of Electrical and Computer Engineering, Faculty of Engineering, Naresuan University, Phitsanulok, 65000, Thailand
– sequence: 5
  givenname: Ivan
  orcidid: 0000-0002-2826-6367
  surname: Lee
  fullname: Lee, Ivan
  email: ivan.lee@unisa.edu.au
  organization: STEM, University of South Australia, Adelaide, SA, 5095, Australia
BookMark eNqFkNtKw0AQhhepYKu-guwLpO4h2SQiaCjrAUq9UK-X6WZWt8SkbLbFvr0p1RtvejXMwPcz_zcho7ZrkZArzqaccXW9mmL7Aes1-Klggg_HVJb5CRnzIpeJylU5ImNWZiLhZa7OyKTvV4wxWaRqTO601lWywHhDq5ZqCM2O6m8fqd52zSb6roWwowvcBGhoFeynj2jjJiB9RRjWC3LqoOnx8neek_cH_TZ7SuYvj8-zap5YyUVMUlDWpQIdc1hznpa4tAIlsFyhQMFFpgonskLWDMFJVyIUkElccpGndW3lObk95NrQ9X1AZ6yPsP8vBvCN4czsZZiV-ZNh9jLMQcaAq3_4Ovivodpx8P4A4lBu6zGY3npsLdY-DCJM3fljET9y-X-H
CitedBy_id crossref_primary_10_1109_JIOT_2025_3565688
crossref_primary_10_3390_machines13060466
crossref_primary_10_1016_j_icte_2024_11_004
crossref_primary_10_1109_ACCESS_2024_3396224
crossref_primary_10_1109_TII_2022_3144016
crossref_primary_10_1007_s11042_022_14187_y
crossref_primary_10_1007_s11220_024_00459_4
crossref_primary_10_1016_j_dsp_2022_103578
crossref_primary_10_1016_j_cviu_2024_104000
crossref_primary_10_3390_electronics13010012
crossref_primary_10_1016_j_asoc_2023_110671
crossref_primary_10_3390_rs16132287
crossref_primary_10_3390_s22010337
crossref_primary_10_1016_j_neucom_2023_01_055
crossref_primary_10_1007_s11042_025_20912_0
crossref_primary_10_1007_s00521_023_08247_z
crossref_primary_10_3390_electronics12030541
crossref_primary_10_1016_j_asoc_2023_110639
crossref_primary_10_1016_j_engappai_2024_109260
crossref_primary_10_1007_s00138_024_01611_6
crossref_primary_10_1016_j_procs_2023_08_178
crossref_primary_10_1109_TIM_2023_3328078
crossref_primary_10_1007_s10489_024_06200_8
crossref_primary_10_3390_sym15112080
crossref_primary_10_1038_s41598_023_36724_x
crossref_primary_10_1016_j_cviu_2022_103561
crossref_primary_10_1109_TII_2022_3153365
crossref_primary_10_1016_j_compeleceng_2022_107864
crossref_primary_10_1109_TIM_2022_3153997
crossref_primary_10_1016_j_engappai_2023_105845
crossref_primary_10_1016_j_asoc_2023_110788
crossref_primary_10_1016_j_neucom_2024_127664
crossref_primary_10_1007_s12065_023_00892_6
crossref_primary_10_1109_TCBB_2023_3272333
crossref_primary_10_1109_ACCESS_2023_3241234
crossref_primary_10_1117_1_JEI_33_5_053063
crossref_primary_10_1016_j_engappai_2024_109030
crossref_primary_10_1109_TIV_2024_3370930
crossref_primary_10_1016_j_eswa_2025_128155
crossref_primary_10_3390_electronics13173487
Cites_doi 10.1145/3321707.3321729
10.1023/A:1022672621406
10.1016/j.neucom.2017.12.049
10.1007/s11263-015-0816-y
10.1016/j.engappai.2017.05.005
10.1016/j.engappai.2019.07.018
10.1007/s11042-019-08332-3
10.1007/978-3-030-01261-8_20
10.1109/4235.996017
10.1007/978-3-030-01264-9_8
10.1145/3065386
10.1007/978-3-030-01246-5_2
10.1007/s11263-009-0275-4
10.1609/aaai.v33i01.33014780
10.1016/j.engappai.2019.08.014
10.1007/978-3-030-01231-1_29
ContentType Journal Article
Copyright 2021 Elsevier Ltd
Copyright_xml – notice: 2021 Elsevier Ltd
DBID AAYXX
CITATION
DOI 10.1016/j.engappai.2021.104397
DatabaseName CrossRef
DatabaseTitle CrossRef
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Applied Sciences
Computer Science
EISSN 1873-6769
ExternalDocumentID 10_1016_j_engappai_2021_104397
S0952197621002451
GroupedDBID --K
--M
.DC
.~1
0R~
1B1
1~.
1~5
29G
4.4
457
4G.
5GY
5VS
7-5
71M
8P~
9JN
AABNK
AACTN
AAEDT
AAEDW
AAIAV
AAIKJ
AAKOC
AALRI
AAOAW
AAQFI
AAQXK
AAXUO
AAYFN
ABBOA
ABMAC
ABXDB
ABYKQ
ACDAQ
ACGFS
ACNNM
ACRLP
ACZNC
ADBBV
ADEZE
ADJOM
ADMUD
ADTZH
AEBSH
AECPX
AEKER
AENEX
AFKWA
AFTJW
AGHFR
AGUBO
AGYEJ
AHHHB
AHJVU
AHZHX
AIALX
AIEXJ
AIKHN
AITUG
AJBFU
AJOXV
ALMA_UNASSIGNED_HOLDINGS
AMFUW
AMRAJ
AOUOD
ASPBG
AVWKF
AXJTR
AZFZN
BJAXD
BKOJK
BLXMC
CS3
DU5
EBS
EFJIC
EFLBG
EJD
EO8
EO9
EP2
EP3
F5P
FDB
FEDTE
FGOYB
FIRID
FNPLU
FYGXN
G-2
G-Q
GBLVA
GBOLZ
HLZ
HVGLF
HZ~
IHE
J1W
JJJVA
KOM
LG9
LY7
M41
MO0
N9A
O-L
O9-
OAUVE
OZT
P-8
P-9
P2P
PC.
Q38
R2-
RIG
ROL
RPZ
SBC
SDF
SDG
SDP
SES
SET
SEW
SPC
SPCBC
SST
SSV
SSZ
T5K
TN5
UHS
WUQ
ZMT
~G-
9DU
AATTM
AAXKI
AAYWO
AAYXX
ABJNI
ABWVN
ACLOT
ACRPL
ACVFH
ADCNI
ADNMO
AEIPS
AEUPX
AFJKZ
AFPUW
AGQPQ
AIGII
AIIUN
AKBMS
AKRWK
AKYEP
ANKPU
APXCP
CITATION
EFKBS
~HD
ID FETCH-LOGICAL-c312t-4a6cf42ef0fed1149ebc2e3a076e2e212568f2583d0eaf3f9ea8a53eb1274ddc3
ISICitedReferencesCount 50
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000686249600001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0952-1976
IngestDate Tue Nov 18 22:00:31 EST 2025
Sat Nov 29 07:07:56 EST 2025
Fri Feb 23 02:43:24 EST 2024
IsPeerReviewed true
IsScholarly true
Keywords Deep learning
Neural Architecture Search
Multi-Objective Evolutionary Algorithms
Image classification
Language English
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c312t-4a6cf42ef0fed1149ebc2e3a076e2e212568f2583d0eaf3f9ea8a53eb1274ddc3
ORCID 0000-0002-2826-6367
0000-0002-1508-3123
ParticipantIDs crossref_citationtrail_10_1016_j_engappai_2021_104397
crossref_primary_10_1016_j_engappai_2021_104397
elsevier_sciencedirect_doi_10_1016_j_engappai_2021_104397
PublicationCentury 2000
PublicationDate September 2021
2021-09-00
PublicationDateYYYYMMDD 2021-09-01
PublicationDate_xml – month: 09
  year: 2021
  text: September 2021
PublicationDecade 2020
PublicationTitle Engineering applications of artificial intelligence
PublicationYear 2021
Publisher Elsevier Ltd
Publisher_xml – name: Elsevier Ltd
References Pham, Guan, Zoph, Le, Dean (b36) 2018
Yu, C., Wang, J., Peng, C., Gao, C., Yu, G., Sang, N., 2018. BiSeNet: Bilateral segmentation network for real-time semantic segmentation. In: Proceedings of the European Conference on Computer Vision (ECCV). pp. 325–341.
Wu, Keutzer, Dai, Zhang, Wang, Sun, Wu, Tian, Vajda, Jia (b49) 2019
Xie, Yuille (b51) 2017
Krizhevsky, Sutskever, Hinton (b25) 2017; 60
Xiao, B., Wu, H., Wei, Y., 2018. Simple baselines for human pose estimation and tracking. In: Proceedings of the European Conference on Computer Vision (ECCV). pp. 472–487.
Tan, Chen, Pang, Vasudevan, Sandler, Howard, Le (b43) 2019
Hasan, Lwin, Imani, Shabut, Bittencourt, Hossain (b17) 2019; 86
Howard, Pang, Adam, Le, Sandler, Chen, Wang, Chen, Tan, Chu, Vasudevan, Zhu (b19) 2019
Szegedy, Liu, Jia, Sermanet, Reed, Anguelov, Erhan, Vanhoucke, Rabinovich (b42) 2015
Ma, N., Zhang, X., Zheng, H.-T., Sun, J., 2018. ShuffleNet V2: Practical guidelines for efficient CNN architecture design. In: Proceedings of the European Conference on Computer Vision (ECCV). pp. 122–138.
Stamoulis, Ding, Wang, Lymberopoulos, Priyantha, Liu, Marculescu (b41) 2019
Real, Moore, Selle, Saxena, Suematsu, Tan, Le, Kurakin (b38) 2017
Chu, Zhang, Xu, Li (b7) 2019
Yu, Peng (b56) 2020
Elsken, Metzen, Hutter (b14) 2018
Sandler, Howard, Zhu, Zhmoginov, Chen (b40) 2018
Real, E., Aggarwal, A., Huang, Y., Le, Q.V., 2019. Regularized evolution for image classifier architecture search. In: Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 33. pp. 4780–4789.
Termritthikun, Jamtsho, Muneesawang (b46) 2020; 79
Han, Wang, Tian, Guo, Xu, Xu (b16) 2020
Li, Talwalkar (b26) 2019
Liu, C., Zoph, B., Neumann, M., Shlens, J., Hua, W., Li, L.-J., Fei-Fei, L., Yuille, A.L., Huang, J., Murphy, K., 2018. Progressive neural architecture search. In: Proceedings of the European Conference on Computer Vision (ECCV). pp. 19–35.
Xu, Xie, Zhang, Chen, Qi, Tian, Xiong (b53) 2020
Zela, Siems, Hutter (b58) 2020
Dong, Yang (b13) 2020
Krizhevsky (b24) 2009
Lin, Maire, Belongie, Hays, Perona, Ramanan, Dollár, Zitnick (b27) 2014
Yang, Wang, Chen, Shi, Xu, Xu, Tian, Xu (b54) 2020
Carrau, Reynoso-Meza, García-Nieto, Blasco (b4) 2017; 64
Hu, Xie, Zheng, Liu, Shi, Liu, Lin (b22) 2020
Iandola, Han, Moskewicz, Ashraf, Dally, Keutzer (b23) 2016
Wan, Dai, Zhang, He, Tian, Xie, Wu, Yu, Xu, Chen, Vajda, Gonzalez (b47) 2020
Dong, Yang (b12) 2019
Termritthikun, Jamtsho, Muneesawang (b45) 2019; 85
Bochkovskiy, Wang, Liao (b2) 2020
Zoph, Le (b60) 2016
Howard, Zhu, Chen, Kalenichenko, Wang, Weyand, Andreetto, Adam (b20) 2017
Williams (b48) 1992; 8
Lu, Deb, Goodman, Banzhaf, Boddeti (b32) 2020
Russakovsky, Deng, Su, Krause, Satheesh, Ma, Huang, Karpathy, Khosla, Bernstein, Berg, Fei-Fei (b39) 2015; 115
Baldominos, Saez, Isasi (b1) 2017; 283
Tan, Le (b44) 2019
Cordts, Omran, Ramos, Rehfeld, Enzweiler, Benenson, Franke, Roth, Schiele (b8) 2016
Xie, Zheng, Liu, Lin (b52) 2018
He, Zhang, Ren, Sun (b18) 2016
Liu, Simonyan, Yang (b30) 2018
Cai, Gan, Wang, Zhang, Han (b3) 2020
DeVries, Taylor (b11) 2017
Liu, Simonyan, Vinyals, Fernando, Kavukcuoglu (b29) 2018
Deb, Pratap, Agarwal, Meyarivan (b10) 2002; 6
Liu, Anguelov, Erhan, Szegedy, Reed, Fu, Berg (b28) 2016
Lu, Z., Whalen, I., Boddeti, V., Dhebar, Y., Deb, K., Goodman, E., Banzhaf, W., 2019. NSGA-Net: neural architecture search using multi-objective genetic algorithm. In: Proceedings of the Genetic and Evolutionary Computation Conference on. pp. 419–427.
Cubuk, Zoph, Mane, Vasudevan, Le (b9) 2019
Hu, Shen, Sun (b21) 2018
Ying, Klein, Christiansen, Real, Murphy, Hutter (b55) 2019
Chu, Zhang, Xu (b6) 2020
Zhang, Zhou, Lin, Sun (b59) 2018
Everingham, Gool, Williams, Winn, Zisserman (b15) 2010; 88
Chen, Xie, Wu, Tian (b5) 2019
Lu, Whalen, Dhebar, Deb, Goodman, Banzhaf, Boddeti (b34) 2020
Krizhevsky (10.1016/j.engappai.2021.104397_b25) 2017; 60
Deb (10.1016/j.engappai.2021.104397_b10) 2002; 6
Zela (10.1016/j.engappai.2021.104397_b58) 2020
Wu (10.1016/j.engappai.2021.104397_b49) 2019
He (10.1016/j.engappai.2021.104397_b18) 2016
Dong (10.1016/j.engappai.2021.104397_b13) 2020
Sandler (10.1016/j.engappai.2021.104397_b40) 2018
Szegedy (10.1016/j.engappai.2021.104397_b42) 2015
Tan (10.1016/j.engappai.2021.104397_b43) 2019
Li (10.1016/j.engappai.2021.104397_b26) 2019
Lin (10.1016/j.engappai.2021.104397_b27) 2014
Williams (10.1016/j.engappai.2021.104397_b48) 1992; 8
Dong (10.1016/j.engappai.2021.104397_b12) 2019
Liu (10.1016/j.engappai.2021.104397_b30) 2018
Ying (10.1016/j.engappai.2021.104397_b55) 2019
Chu (10.1016/j.engappai.2021.104397_b6) 2020
Xu (10.1016/j.engappai.2021.104397_b53) 2020
Yu (10.1016/j.engappai.2021.104397_b56) 2020
Howard (10.1016/j.engappai.2021.104397_b19) 2019
Chu (10.1016/j.engappai.2021.104397_b7) 2019
10.1016/j.engappai.2021.104397_b50
DeVries (10.1016/j.engappai.2021.104397_b11) 2017
Wan (10.1016/j.engappai.2021.104397_b47) 2020
10.1016/j.engappai.2021.104397_b57
Russakovsky (10.1016/j.engappai.2021.104397_b39) 2015; 115
Bochkovskiy (10.1016/j.engappai.2021.104397_b2) 2020
Hu (10.1016/j.engappai.2021.104397_b22) 2020
Stamoulis (10.1016/j.engappai.2021.104397_b41) 2019
Real (10.1016/j.engappai.2021.104397_b38) 2017
Everingham (10.1016/j.engappai.2021.104397_b15) 2010; 88
Cubuk (10.1016/j.engappai.2021.104397_b9) 2019
Zhang (10.1016/j.engappai.2021.104397_b59) 2018
Liu (10.1016/j.engappai.2021.104397_b28) 2016
Lu (10.1016/j.engappai.2021.104397_b32) 2020
Howard (10.1016/j.engappai.2021.104397_b20) 2017
Xie (10.1016/j.engappai.2021.104397_b52) 2018
Tan (10.1016/j.engappai.2021.104397_b44) 2019
Termritthikun (10.1016/j.engappai.2021.104397_b46) 2020; 79
Iandola (10.1016/j.engappai.2021.104397_b23) 2016
Han (10.1016/j.engappai.2021.104397_b16) 2020
Krizhevsky (10.1016/j.engappai.2021.104397_b24) 2009
Pham (10.1016/j.engappai.2021.104397_b36) 2018
Elsken (10.1016/j.engappai.2021.104397_b14) 2018
Chen (10.1016/j.engappai.2021.104397_b5) 2019
Cai (10.1016/j.engappai.2021.104397_b3) 2020
10.1016/j.engappai.2021.104397_b37
Xie (10.1016/j.engappai.2021.104397_b51) 2017
Carrau (10.1016/j.engappai.2021.104397_b4) 2017; 64
Zoph (10.1016/j.engappai.2021.104397_b60) 2016
Cordts (10.1016/j.engappai.2021.104397_b8) 2016
10.1016/j.engappai.2021.104397_b31
Baldominos (10.1016/j.engappai.2021.104397_b1) 2017; 283
10.1016/j.engappai.2021.104397_b33
10.1016/j.engappai.2021.104397_b35
Liu (10.1016/j.engappai.2021.104397_b29) 2018
Lu (10.1016/j.engappai.2021.104397_b34) 2020
Termritthikun (10.1016/j.engappai.2021.104397_b45) 2019; 85
Hasan (10.1016/j.engappai.2021.104397_b17) 2019; 86
Hu (10.1016/j.engappai.2021.104397_b21) 2018
Yang (10.1016/j.engappai.2021.104397_b54) 2020
References_xml – year: 2020
  ident: b2
  article-title: Yolov4: Optimal speed and accuracy of object detection
– year: 2018
  ident: b30
  article-title: Darts: Differentiable architecture search
  publication-title: International Conference on Learning Representations
– start-page: 2820
  year: 2019
  end-page: 2828
  ident: b43
  article-title: Mnasnet: Platform-aware neural architecture search for mobile
  publication-title: 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
– year: 2018
  ident: b29
  article-title: Hierarchical representations for efficient architecture search
  publication-title: International Conference on Learning Representations
– year: 2017
  ident: b11
  article-title: Improved regularization of convolutional neural networks with cutout
– reference: Liu, C., Zoph, B., Neumann, M., Shlens, J., Hua, W., Li, L.-J., Fei-Fei, L., Yuille, A.L., Huang, J., Murphy, K., 2018. Progressive neural architecture search. In: Proceedings of the European Conference on Computer Vision (ECCV). pp. 19–35.
– start-page: 3213
  year: 2016
  end-page: 3223
  ident: b8
  article-title: The cityscapes dataset for semantic urban scene understanding
  publication-title: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
– start-page: 1
  year: 2020
  ident: b34
  article-title: Multi-objective evolutionary design of deep convolutional neural networks for image classification
  publication-title: IEEE Trans. Evol. Comput.
– reference: Ma, N., Zhang, X., Zheng, H.-T., Sun, J., 2018. ShuffleNet V2: Practical guidelines for efficient CNN architecture design. In: Proceedings of the European Conference on Computer Vision (ECCV). pp. 122–138.
– start-page: 4510
  year: 2018
  end-page: 4520
  ident: b40
  article-title: MobileNetV2: Inverted residuals and linear bottlenecks
  publication-title: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
– year: 2018
  ident: b36
  article-title: Efficient neural architecture search via parameter sharing
– reference: Lu, Z., Whalen, I., Boddeti, V., Dhebar, Y., Deb, K., Goodman, E., Banzhaf, W., 2019. NSGA-Net: neural architecture search using multi-objective genetic algorithm. In: Proceedings of the Genetic and Evolutionary Computation Conference on. pp. 419–427.
– volume: 64
  start-page: 52
  year: 2017
  end-page: 66
  ident: b4
  article-title: Enhancing controller’s tuning reliability with multi-objective optimisation: From model in the loop to hardware in the loop
  publication-title: Eng. Appl. Artif. Intell.
– volume: 86
  start-page: 107
  year: 2019
  end-page: 135
  ident: b17
  article-title: Dynamic multi-objective optimisation using deep reinforcement learning: benchmark, algorithm and an application to identify vulnerable zones based on water quality
  publication-title: Eng. Appl. Artif. Intell.
– volume: 6
  start-page: 182
  year: 2002
  end-page: 197
  ident: b10
  article-title: A fast and elitist multiobjective genetic algorithm: NSGA-II
  publication-title: IEEE Trans. Evol. Comput.
– year: 2020
  ident: b13
  article-title: Nas-bench-201: Extending the scope of reproducible neural architecture search
  publication-title: Eighth International Conference on Learning Representations
– start-page: 35
  year: 2020
  end-page: 51
  ident: b32
  article-title: NSGANetV2: Evolutionary multi-objective surrogate-assisted neural architecture search
  publication-title: European Conference on Computer Vision
– start-page: 740
  year: 2014
  end-page: 755
  ident: b27
  article-title: Microsoft COCO: Common objects in context
  publication-title: European Conference on Computer Vision
– start-page: 770
  year: 2016
  end-page: 778
  ident: b18
  article-title: Deep residual learning for image recognition
  publication-title: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
– start-page: 7132
  year: 2018
  end-page: 7141
  ident: b21
  article-title: Squeeze-and-excitation networks
  publication-title: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
– year: 2019
  ident: b7
  article-title: Fairnas: Rethinking evaluation fairness of weight sharing neural architecture search
– year: 2016
  ident: b60
  article-title: Neural architecture search with reinforcement learning
  publication-title: ICLR
– volume: 88
  start-page: 303
  year: 2010
  end-page: 338
  ident: b15
  article-title: The pascal visual object classes (VOC) challenge
  publication-title: Int. J. Comput. Vis.
– start-page: 7105
  year: 2019
  end-page: 7114
  ident: b55
  article-title: Nas-bench-101: Towards reproducible neural architecture search
  publication-title: International Conference on Machine Learning
– year: 2018
  ident: b14
  article-title: Efficient multi-objective neural architecture search via lamarckian evolution
  publication-title: International Conference on Learning Representations
– year: 2016
  ident: b23
  article-title: Squeezenet: Alexnet-level accuracy with 50x fewer parameters and <0.5mb model size
– start-page: 10734
  year: 2019
  end-page: 10742
  ident: b49
  article-title: Fbnet: Hardware-aware efficient convnet design via differentiable neural architecture search
  publication-title: 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
– volume: 283
  start-page: 38
  year: 2017
  end-page: 52
  ident: b1
  article-title: Evolutionary convolutional neural networks: An application to handwriting recognition
  publication-title: Neurocomputing
– year: 2020
  ident: b3
  article-title: Once for all: Train one network and specialize it for efficient deployment
  publication-title: ICLR 2020 : Eighth International Conference on Learning Representations
– volume: 115
  start-page: 211
  year: 2015
  end-page: 252
  ident: b39
  article-title: Imagenet large scale visual recognition challenge
  publication-title: Int. J. Comput. Vis.
– year: 2020
  ident: b53
  article-title: Pc-DARTS: Partial channel connections for memory-efficient architecture search
  publication-title: ICLR 2020 : Eighth International Conference on Learning Representations
– start-page: 1294
  year: 2019
  end-page: 1303
  ident: b5
  article-title: Progressive differentiable architecture search: Bridging the depth gap between search and evaluation
  publication-title: 2019 IEEE/CVF International Conference on Computer Vision (ICCV)
– start-page: 1314
  year: 2019
  end-page: 1324
  ident: b19
  article-title: Searching for MobileNetV3
  publication-title: 2019 IEEE/CVF International Conference on Computer Vision (ICCV)
– year: 2020
  ident: b56
  article-title: Cyclic differentiable architecture search
– start-page: 1580
  year: 2020
  end-page: 1589
  ident: b16
  article-title: Ghostnet: More features from cheap operations
  publication-title: 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
– start-page: 1
  year: 2015
  end-page: 9
  ident: b42
  article-title: Going deeper with convolutions
  publication-title: 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
– year: 2020
  ident: b58
  article-title: Nas-bench-1shot1: Benchmarking and dissecting one-shot neural architecture search
  publication-title: ICLR 2020 : Eighth International Conference on Learning Representations
– start-page: 2902
  year: 2017
  end-page: 2911
  ident: b38
  article-title: Large-scale evolution of image classifiers
  publication-title: ICML’17 Proceedings of the 34th International Conference on Machine Learning - Volume 70
– start-page: 113
  year: 2019
  end-page: 123
  ident: b9
  article-title: Autoaugment: Learning augmentation strategies from data
  publication-title: 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
– year: 2017
  ident: b51
  article-title: Genetic CNN
  publication-title: 2017 IEEE International Conference on Computer Vision (ICCV)
– start-page: 367
  year: 2019
  end-page: 377
  ident: b26
  article-title: Random search and reproducibility for neural architecture search
  publication-title: Uncertainty in Artificial Intelligence
– year: 2017
  ident: b20
  article-title: Mobilenets: Efficient convolutional neural networks for mobile vision applications
– reference: Yu, C., Wang, J., Peng, C., Gao, C., Yu, G., Sang, N., 2018. BiSeNet: Bilateral segmentation network for real-time semantic segmentation. In: Proceedings of the European Conference on Computer Vision (ECCV). pp. 325–341.
– year: 2009
  ident: b24
  article-title: Learning multiple layers of features from tiny images
– volume: 85
  start-page: 579
  year: 2019
  end-page: 589
  ident: b45
  article-title: On-device facial verification using NUF-net model of deep learning
  publication-title: Eng. Appl. Artif. Intell.
– volume: 79
  start-page: 1475
  year: 2020
  end-page: 1495
  ident: b46
  article-title: An improved residual network model for image recognition using a combination of snapshot ensembles and the cutout technique
  publication-title: Multimedia Tools Appl.
– start-page: 12084
  year: 2020
  end-page: 12092
  ident: b22
  article-title: Dsnas: Direct neural architecture search without parameter retraining
  publication-title: 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
– start-page: 1761
  year: 2019
  end-page: 1770
  ident: b12
  article-title: Searching for a robust neural architecture in four GPU hours
  publication-title: 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
– reference: Real, E., Aggarwal, A., Huang, Y., Le, Q.V., 2019. Regularized evolution for image classifier architecture search. In: Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 33. pp. 4780–4789.
– start-page: 481
  year: 2019
  end-page: 497
  ident: b41
  article-title: Single-path NAS: Designing hardware-efficient convnets in less than 4 hours
  publication-title: Joint European Conference on Machine Learning and Knowledge Discovery in Databases
– start-page: 12965
  year: 2020
  end-page: 12974
  ident: b47
  article-title: FBNetV2: Differentiable neural architecture search for spatial and channel dimensions
  publication-title: 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
– start-page: 6105
  year: 2019
  end-page: 6114
  ident: b44
  article-title: Efficientnet: Rethinking model scaling for convolutional neural networks
  publication-title: International Conference on Machine Learning
– volume: 60
  start-page: 84
  year: 2017
  end-page: 90
  ident: b25
  article-title: Imagenet classification with deep convolutional neural networks
  publication-title: Commun. ACM
– year: 2018
  ident: b52
  article-title: Snas: stochastic neural architecture search
  publication-title: International Conference on Learning Representations
– start-page: 21
  year: 2016
  end-page: 37
  ident: b28
  article-title: Ssd: Single shot multibox detector
  publication-title: 14th European Conference on Computer Vision, ECCV 2016
– start-page: 4042
  year: 2020
  end-page: 4046
  ident: b6
  article-title: Moga: Searching beyond Mobilenetv3
  publication-title: ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
– start-page: 1829
  year: 2020
  end-page: 1838
  ident: b54
  article-title: Cars: Continuous evolution for efficient neural architecture search
  publication-title: 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
– start-page: 6848
  year: 2018
  end-page: 6856
  ident: b59
  article-title: Shufflenet: An extremely efficient convolutional neural network for mobile devices
  publication-title: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
– volume: 8
  start-page: 229
  year: 1992
  end-page: 256
  ident: b48
  article-title: Simple statistical gradient-following algorithms for connectionist reinforcement learning
  publication-title: Mach. Learn.
– reference: Xiao, B., Wu, H., Wei, Y., 2018. Simple baselines for human pose estimation and tracking. In: Proceedings of the European Conference on Computer Vision (ECCV). pp. 472–487.
– ident: 10.1016/j.engappai.2021.104397_b33
  doi: 10.1145/3321707.3321729
– year: 2018
  ident: 10.1016/j.engappai.2021.104397_b52
  article-title: Snas: stochastic neural architecture search
– year: 2017
  ident: 10.1016/j.engappai.2021.104397_b11
– year: 2020
  ident: 10.1016/j.engappai.2021.104397_b53
  article-title: Pc-DARTS: Partial channel connections for memory-efficient architecture search
– volume: 8
  start-page: 229
  issue: 3
  year: 1992
  ident: 10.1016/j.engappai.2021.104397_b48
  article-title: Simple statistical gradient-following algorithms for connectionist reinforcement learning
  publication-title: Mach. Learn.
  doi: 10.1023/A:1022672621406
– volume: 283
  start-page: 38
  year: 2017
  ident: 10.1016/j.engappai.2021.104397_b1
  article-title: Evolutionary convolutional neural networks: An application to handwriting recognition
  publication-title: Neurocomputing
  doi: 10.1016/j.neucom.2017.12.049
– year: 2016
  ident: 10.1016/j.engappai.2021.104397_b60
  article-title: Neural architecture search with reinforcement learning
– start-page: 6848
  year: 2018
  ident: 10.1016/j.engappai.2021.104397_b59
  article-title: Shufflenet: An extremely efficient convolutional neural network for mobile devices
– year: 2018
  ident: 10.1016/j.engappai.2021.104397_b30
  article-title: Darts: Differentiable architecture search
– start-page: 1761
  year: 2019
  ident: 10.1016/j.engappai.2021.104397_b12
  article-title: Searching for a robust neural architecture in four GPU hours
– start-page: 4042
  year: 2020
  ident: 10.1016/j.engappai.2021.104397_b6
  article-title: Moga: Searching beyond Mobilenetv3
– start-page: 3213
  year: 2016
  ident: 10.1016/j.engappai.2021.104397_b8
  article-title: The cityscapes dataset for semantic urban scene understanding
– volume: 115
  start-page: 211
  issue: 3
  year: 2015
  ident: 10.1016/j.engappai.2021.104397_b39
  article-title: Imagenet large scale visual recognition challenge
  publication-title: Int. J. Comput. Vis.
  doi: 10.1007/s11263-015-0816-y
– year: 2018
  ident: 10.1016/j.engappai.2021.104397_b14
  article-title: Efficient multi-objective neural architecture search via lamarckian evolution
– start-page: 10734
  year: 2019
  ident: 10.1016/j.engappai.2021.104397_b49
  article-title: Fbnet: Hardware-aware efficient convnet design via differentiable neural architecture search
– year: 2019
  ident: 10.1016/j.engappai.2021.104397_b7
– start-page: 740
  year: 2014
  ident: 10.1016/j.engappai.2021.104397_b27
  article-title: Microsoft COCO: Common objects in context
– volume: 64
  start-page: 52
  year: 2017
  ident: 10.1016/j.engappai.2021.104397_b4
  article-title: Enhancing controller’s tuning reliability with multi-objective optimisation: From model in the loop to hardware in the loop
  publication-title: Eng. Appl. Artif. Intell.
  doi: 10.1016/j.engappai.2017.05.005
– year: 2017
  ident: 10.1016/j.engappai.2021.104397_b51
  article-title: Genetic CNN
– volume: 85
  start-page: 579
  year: 2019
  ident: 10.1016/j.engappai.2021.104397_b45
  article-title: On-device facial verification using NUF-net model of deep learning
  publication-title: Eng. Appl. Artif. Intell.
  doi: 10.1016/j.engappai.2019.07.018
– volume: 79
  start-page: 1475
  issue: 1
  year: 2020
  ident: 10.1016/j.engappai.2021.104397_b46
  article-title: An improved residual network model for image recognition using a combination of snapshot ensembles and the cutout technique
  publication-title: Multimedia Tools Appl.
  doi: 10.1007/s11042-019-08332-3
– year: 2020
  ident: 10.1016/j.engappai.2021.104397_b2
– start-page: 35
  year: 2020
  ident: 10.1016/j.engappai.2021.104397_b32
  article-title: NSGANetV2: Evolutionary multi-objective surrogate-assisted neural architecture search
– start-page: 7105
  year: 2019
  ident: 10.1016/j.engappai.2021.104397_b55
  article-title: Nas-bench-101: Towards reproducible neural architecture search
– start-page: 12965
  year: 2020
  ident: 10.1016/j.engappai.2021.104397_b47
  article-title: FBNetV2: Differentiable neural architecture search for spatial and channel dimensions
– start-page: 1294
  year: 2019
  ident: 10.1016/j.engappai.2021.104397_b5
  article-title: Progressive differentiable architecture search: Bridging the depth gap between search and evaluation
– year: 2020
  ident: 10.1016/j.engappai.2021.104397_b3
  article-title: Once for all: Train one network and specialize it for efficient deployment
– start-page: 7132
  year: 2018
  ident: 10.1016/j.engappai.2021.104397_b21
  article-title: Squeeze-and-excitation networks
– ident: 10.1016/j.engappai.2021.104397_b57
  doi: 10.1007/978-3-030-01261-8_20
– start-page: 1580
  year: 2020
  ident: 10.1016/j.engappai.2021.104397_b16
  article-title: Ghostnet: More features from cheap operations
– start-page: 770
  year: 2016
  ident: 10.1016/j.engappai.2021.104397_b18
  article-title: Deep residual learning for image recognition
– start-page: 2820
  year: 2019
  ident: 10.1016/j.engappai.2021.104397_b43
  article-title: Mnasnet: Platform-aware neural architecture search for mobile
– start-page: 481
  year: 2019
  ident: 10.1016/j.engappai.2021.104397_b41
  article-title: Single-path NAS: Designing hardware-efficient convnets in less than 4 hours
– year: 2020
  ident: 10.1016/j.engappai.2021.104397_b58
  article-title: Nas-bench-1shot1: Benchmarking and dissecting one-shot neural architecture search
– year: 2020
  ident: 10.1016/j.engappai.2021.104397_b13
  article-title: Nas-bench-201: Extending the scope of reproducible neural architecture search
– volume: 6
  start-page: 182
  issue: 2
  year: 2002
  ident: 10.1016/j.engappai.2021.104397_b10
  article-title: A fast and elitist multiobjective genetic algorithm: NSGA-II
  publication-title: IEEE Trans. Evol. Comput.
  doi: 10.1109/4235.996017
– start-page: 2902
  year: 2017
  ident: 10.1016/j.engappai.2021.104397_b38
  article-title: Large-scale evolution of image classifiers
– start-page: 6105
  year: 2019
  ident: 10.1016/j.engappai.2021.104397_b44
  article-title: Efficientnet: Rethinking model scaling for convolutional neural networks
– start-page: 4510
  year: 2018
  ident: 10.1016/j.engappai.2021.104397_b40
  article-title: MobileNetV2: Inverted residuals and linear bottlenecks
– ident: 10.1016/j.engappai.2021.104397_b35
  doi: 10.1007/978-3-030-01264-9_8
– volume: 60
  start-page: 84
  issue: 6
  year: 2017
  ident: 10.1016/j.engappai.2021.104397_b25
  article-title: Imagenet classification with deep convolutional neural networks
  publication-title: Commun. ACM
  doi: 10.1145/3065386
– start-page: 1
  year: 2015
  ident: 10.1016/j.engappai.2021.104397_b42
  article-title: Going deeper with convolutions
– ident: 10.1016/j.engappai.2021.104397_b31
  doi: 10.1007/978-3-030-01246-5_2
– volume: 88
  start-page: 303
  issue: 2
  year: 2010
  ident: 10.1016/j.engappai.2021.104397_b15
  article-title: The pascal visual object classes (VOC) challenge
  publication-title: Int. J. Comput. Vis.
  doi: 10.1007/s11263-009-0275-4
– year: 2020
  ident: 10.1016/j.engappai.2021.104397_b56
– ident: 10.1016/j.engappai.2021.104397_b37
  doi: 10.1609/aaai.v33i01.33014780
– start-page: 1
  year: 2020
  ident: 10.1016/j.engappai.2021.104397_b34
  article-title: Multi-objective evolutionary design of deep convolutional neural networks for image classification
  publication-title: IEEE Trans. Evol. Comput.
– year: 2009
  ident: 10.1016/j.engappai.2021.104397_b24
– start-page: 1314
  year: 2019
  ident: 10.1016/j.engappai.2021.104397_b19
  article-title: Searching for MobileNetV3
– start-page: 113
  year: 2019
  ident: 10.1016/j.engappai.2021.104397_b9
  article-title: Autoaugment: Learning augmentation strategies from data
– volume: 86
  start-page: 107
  year: 2019
  ident: 10.1016/j.engappai.2021.104397_b17
  article-title: Dynamic multi-objective optimisation using deep reinforcement learning: benchmark, algorithm and an application to identify vulnerable zones based on water quality
  publication-title: Eng. Appl. Artif. Intell.
  doi: 10.1016/j.engappai.2019.08.014
– start-page: 21
  year: 2016
  ident: 10.1016/j.engappai.2021.104397_b28
  article-title: Ssd: Single shot multibox detector
– start-page: 1829
  year: 2020
  ident: 10.1016/j.engappai.2021.104397_b54
  article-title: Cars: Continuous evolution for efficient neural architecture search
– start-page: 367
  year: 2019
  ident: 10.1016/j.engappai.2021.104397_b26
  article-title: Random search and reproducibility for neural architecture search
– year: 2017
  ident: 10.1016/j.engappai.2021.104397_b20
– year: 2018
  ident: 10.1016/j.engappai.2021.104397_b36
– start-page: 12084
  year: 2020
  ident: 10.1016/j.engappai.2021.104397_b22
  article-title: Dsnas: Direct neural architecture search without parameter retraining
– year: 2016
  ident: 10.1016/j.engappai.2021.104397_b23
– ident: 10.1016/j.engappai.2021.104397_b50
  doi: 10.1007/978-3-030-01231-1_29
– year: 2018
  ident: 10.1016/j.engappai.2021.104397_b29
  article-title: Hierarchical representations for efficient architecture search
SSID ssj0003846
Score 2.543925
Snippet The goals of this research were to search for Convolutional Neural Network (CNN) architectures, suitable for an on-device processor with limited computing...
SourceID crossref
elsevier
SourceType Enrichment Source
Index Database
Publisher
StartPage 104397
SubjectTerms Deep learning
Image classification
Multi-Objective Evolutionary Algorithms
Neural Architecture Search
Title EEEA-Net: An Early Exit Evolutionary Neural Architecture Search
URI https://dx.doi.org/10.1016/j.engappai.2021.104397
Volume 104
WOSCitedRecordID wos000686249600001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVESC
  databaseName: Elsevier SD Freedom Collection Journals 2021
  customDbUrl:
  eissn: 1873-6769
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0003846
  issn: 0952-1976
  databaseCode: AIEXJ
  dateStart: 19950201
  isFulltext: true
  titleUrlDefault: https://www.sciencedirect.com
  providerName: Elsevier
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1Lb5wwELa2SQ-99F0lfcmH3iK2YAO2c4lW1UZpDqtKTaXtCXnB7pJNaARku_8kfzdjjIFWq6Y59ILQSB4D82mYGc8DoQ-xJGFMg9RjXEgvJMz3FnShPZpxsmC-kEL7zbAJNpvx-Vx8GY1uXC3M-oIVBd9sxNV_FTXQQNimdPYe4u6YAgHuQehwBbHD9Z8ED97bxJspF_KzHYynm7w-mK7bjU2mnOnKYcQzPEj42ke2XLC-b1d4MDzrbtIHyibPqJn6MWjs2cUCQOeXeV0v89W1O9lfrYDSJe3Iy7paNqHa76pa5h1KlbyspLQ596d5aZqT97iAp6nkrzbKbQ6-ZFkMQxck6HKz-hgk8QJhB8B06tgPBwo1MKW7bKuut2GH87EqfsAHkPnYbDHuF_zeXPuPn16Xiuiy3M4TxycxfBLL5wHaJSwSoPF3J5-n89PuJ0-5rQFzbzAoPt_-RNvtnoEtc_YUPW6dEDyx4HmGRqp4jp60Dglu1X0FJDfzw9FeoCMHr0M8KXADLmzAhYfgwhZceAgubMH1En07np59OvHaIRxeSgNSe6GMUx0SpX2tMnCehVqkRFHps1gRBYZPFHNNIk4zX0lNtVCSy4iCCUBYmGUpfYV2ip-F2kMYaDzzM7DQFTgSPhfAhgimNNGgFVK2jyL3iZK07VBvBqVcJH8X0j762K27sj1a7lwhnASS1tK0FmQC4Lpj7et77_YGPerR_xbt1OW1eocepus6r8r3LbJuAdLapCs
linkProvider Elsevier
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=EEEA-Net%3A+An+Early+Exit+Evolutionary+Neural+Architecture+Search&rft.jtitle=Engineering+applications+of+artificial+intelligence&rft.au=Termritthikun%2C+Chakkrit&rft.au=Jamtsho%2C+Yeshi&rft.au=Ieamsaard%2C+Jirarat&rft.au=Muneesawang%2C+Paisarn&rft.date=2021-09-01&rft.issn=0952-1976&rft.volume=104&rft.spage=104397&rft_id=info:doi/10.1016%2Fj.engappai.2021.104397&rft.externalDBID=n%2Fa&rft.externalDocID=10_1016_j_engappai_2021_104397
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0952-1976&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0952-1976&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0952-1976&client=summon