{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,3]],"date-time":"2026-03-03T16:22:05Z","timestamp":1772554925574,"version":"3.50.1"},"reference-count":61,"publisher":"Springer Science and Business Media LLC","issue":"3","license":[{"start":{"date-parts":[[2025,3,13]],"date-time":"2025-03-13T00:00:00Z","timestamp":1741824000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,3,13]],"date-time":"2025-03-13T00:00:00Z","timestamp":1741824000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/100010661","name":"Horizon 2020 Framework Programme","doi-asserted-by":"publisher","award":["739551"],"award-info":[{"award-number":["739551"]}],"id":[{"id":"10.13039\/100010661","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100012996","name":"University of Cyprus","doi-asserted-by":"crossref","id":[{"id":"10.13039\/100012996","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["SN COMPUT. SCI."],"abstract":"<jats:title>Abstract<\/jats:title>\n          <jats:p>Neural architecture search (NAS) has shown promise towards automating neural network design for a given task, but it is computationally demanding due to training costs associated with evaluating a large number of architectures to find the optimal one. To speed up NAS, recent works limit the search to network building blocks (modular search) instead of searching the entire architecture (global search), approximate candidates\u2019 performance evaluation in lieu of complete training, and use gradient descent rather than naturally suitable discrete optimization approaches. However, modular search does not determine network\u2019s macro architecture i.e. depth and width, demanding manual trial and error post-search, hence lacking automation. In this work, we revisit NAS and design a navigable, yet architecturally diverse, macro-micro search space. In addition, to determine relative rankings of candidates, existing methods employ consistent approximations across entire search spaces, whereas different networks may not be fairly comparable under one training protocol. Hence, we propose an architecture-aware approximation with variable training schemes for different networks. Moreover, we develop an efficient search strategy by disjoining macro-micro network design that yields competitive architectures in terms of both accuracy and size. Our proposed framework achieves a new state-of-the-art on EMNIST and KMNIST, while being highly competitive on the CIFAR-10, CIFAR-100, and FashionMNIST datasets and being 2-4 times faster than the fastest global search methods. Lastly, we demonstrate the transferability of our framework to real-world computer vision problems by discovering competitive architectures for face recognition applications.<\/jats:p>","DOI":"10.1007\/s42979-025-03790-z","type":"journal-article","created":{"date-parts":[[2025,3,13]],"date-time":"2025-03-13T05:34:28Z","timestamp":1741844068000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Efficient Global Neural Architecture Search"],"prefix":"10.1007","volume":"6","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6176-2952","authenticated-orcid":false,"given":"Shahid","family":"Siddiqui","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7926-7642","authenticated-orcid":false,"given":"Christos","family":"Kyrkou","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8727-9152","authenticated-orcid":false,"given":"Theocharis","family":"Theocharides","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"297","published-online":{"date-parts":[[2025,3,13]]},"reference":[{"key":"3790_CR1","unstructured":"Simonyan Karen, Zisserman Andrew. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556 (2014)."},{"key":"3790_CR2","doi-asserted-by":"crossref","unstructured":"Long Jonathan, Shelhamer Evan, Darrell Trevor. Fully convolutional networks for semantic segmentation. Proceedings of the IEEE conference on computer vision and pattern recognition. 2015.","DOI":"10.1109\/CVPR.2015.7298965"},{"key":"3790_CR3","doi-asserted-by":"crossref","unstructured":"He Kaiming et al. Deep residual learning for image recognition. Proceedings of the IEEE conference on computer vision and pattern recognition. 2016.","DOI":"10.1109\/CVPR.2016.90"},{"key":"3790_CR4","unstructured":"Zoph Barret, Le Quoc V. Neural architecture search with reinforcement learning. arXiv preprint arXiv:1611.01578 (2016)."},{"key":"3790_CR5","unstructured":"Baker Bowen et al. Designing neural network architectures using reinforcement learning. arXiv preprint arXiv:1611.02167 2016."},{"key":"3790_CR6","unstructured":"Real Esteban et al. Large-scale evolution of image classifiers. International Conference on Machine Learning. PMLR, 2017."},{"key":"3790_CR7","doi-asserted-by":"crossref","unstructured":"Suganuma Masanori, Shirakawa Shinichi, Nagao Tomoharu. A genetic programming approach to designing convolutional neural network architectures. Proceedings of the genetic and evolutionary computation conference. 2017.","DOI":"10.1145\/3071178.3071229"},{"key":"3790_CR8","unstructured":"Elsken Thomas, Metzen Jan-Hendrik, Hutter Frank. Simple and efficient architecture search for convolutional neural networks. arXiv preprint arXiv:1711.04528 2017."},{"key":"3790_CR9","doi-asserted-by":"crossref","unstructured":"Zoph Barret et al. Learning transferable architectures for scalable image recognition. Proceedings of the IEEE conference on computer vision and pattern recognition. 2018.","DOI":"10.1109\/CVPR.2018.00907"},{"key":"3790_CR10","doi-asserted-by":"crossref","unstructured":"Cai Han et al. Efficient architecture search by network transformation. Proceedings of the AAAI Conference on Artificial Intelligence. 2018;32(1).","DOI":"10.1609\/aaai.v32i1.11709"},{"key":"3790_CR11","unstructured":"Kandasamy Kirthevasan et al. Neural architecture search with bayesian optimisation and optimal transport. Adv Neural Inf Proc Syst 2018;31."},{"key":"3790_CR12","doi-asserted-by":"crossref","unstructured":"Dong Jin-Dong et al. Dpp-net: Device-aware progressive search for pareto-optimal neural architectures. Proceedings of the European Conference on Computer Vision (ECCV). 2018.","DOI":"10.1007\/978-3-030-01252-6_32"},{"key":"3790_CR13","unstructured":"Hu Hanzhang et al. Macro neural architecture search revisited. 2nd Workshop on Meta-Learning at NeurIPS. 2018."},{"key":"3790_CR14","unstructured":"Pham H, Guan M, Zoph B, Le Q, Dean J. Efficient neural architecture search via parameters sharing. InInternational conference on machine learning 2018;4095-4104."},{"key":"3790_CR15","unstructured":"Liu Hanxiao, Simonyan Karen, Yang Yiming. Darts: Differentiable architecture search. arXiv preprint arXiv:1806.09055 (2018)."},{"key":"3790_CR16","doi-asserted-by":"crossref","unstructured":"Dong Xuanyi, Yang Yi. Searching for a robust neural architecture in four gpu hours. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition. 2019.","DOI":"10.1109\/CVPR.2019.00186"},{"key":"3790_CR17","unstructured":"Hu Hanzhang et al. \"Efficient forward architecture search.\" Adv Neural Inf Proc Syst 32 (2019)."},{"key":"3790_CR18","unstructured":"Yang Antoine, Pedro M. Esperan\u00e7a, and Fabio M. Carlucci. NAS evaluation is frustratingly hard. arXiv preprint arXiv:1912.12522 (2019)."},{"key":"3790_CR19","doi-asserted-by":"crossref","unstructured":"Lu Zhichao et al. Nsga-net: neural architecture search using multi-objective genetic algorithm. Proceedings of the genetic and evolutionary computation conference. 2019.","DOI":"10.1145\/3321707.3321729"},{"issue":"1","key":"3790_CR20","first-page":"9820","volume":"21","author":"Marius Lindauer","year":"2020","unstructured":"Lindauer Marius, Hutter Frank. Best practices for scientific research on neural architecture search. J Machi Learn Res. 2020;21(1):9820\u201337.","journal-title":"J Machi Learn Res"},{"issue":"4","key":"3790_CR21","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3447582","volume":"54","author":"Pengzhen Ren","year":"2021","unstructured":"Ren Pengzhen, et al. A comprehensive survey of neural architecture search: challenges and solutions. ACM Comput Surveys (CSUR). 2021;54(4):1\u201334.","journal-title":"ACM Comput Surveys (CSUR)"},{"key":"3790_CR22","first-page":"28454","volume":"34","author":"Colin White","year":"2021","unstructured":"White Colin, et al. How powerful are performance predictors in neural architecture search? Adv Neural Inf Proc Syst. 2021;34:28454\u201369.","journal-title":"Adv Neural Inf Proc Syst"},{"key":"3790_CR23","unstructured":"Sun Zihao et al. \u201cAGNAS: Attention-Guided Micro and Macro-Architecture Search.\u201d International Conference on Machine Learning. PMLR, 2022."},{"key":"3790_CR24","unstructured":"Lopes Vasco, Lu\u00eds A. Alexandre. \"Towards Less Constrained Macro-Neural Architecture Search.\" arXiv preprint arXiv:2203.05508 (2022)."},{"key":"3790_CR25","doi-asserted-by":"crossref","unstructured":"Sinha Nilotpal, Chen Kuan-Wen. \u201cNeural architecture search using progressive evolution.\u201d Proc Gen Evolut Comput Conf. 2022.","DOI":"10.1145\/3512290.3528707"},{"key":"3790_CR26","unstructured":"Yu Kaicheng, Sciuto Christian, Jaggi Martin, Musat Claudiu, Salzmann Mathieu. Evaluating the search phase of neural architecture search. In International Conference on Learning Representations, 2019."},{"key":"3790_CR27","unstructured":"White Colin, Safari Mahmoud, Sukthanker Rhea, Ru Binxin, Elsken Thomas, Zela Arber, Dey Debadeepta, Hutter Frank. \u201cNeural architecture search: Insights from 1000 papers.\u201d arXiv preprint arXiv:2301.08727 2023."},{"issue":"35","key":"3790_CR28","first-page":"38788","volume":"6","author":"A Bansal","year":"2022","unstructured":"Bansal A, Stoll D, Janowski M, Zela A, Hutter F. Jahs-bench-201: a foundation for research on joint architecture and hyperparameter search. Adv Neural Inf Proc Syst. 2022;6(35):38788\u2013802.","journal-title":"Adv Neural Inf Proc Syst"},{"key":"3790_CR29","unstructured":"Baker B, Gupta O, Raskar R, Naik N. Accelerating neural architecture search using performance prediction. arXiv preprint arXiv:1705.10823. 2017."},{"key":"3790_CR30","doi-asserted-by":"crossref","unstructured":"Zheng X, Ji R, Tang L, Zhang B, Liu J, Tian Q. Multinomial distribution learning for effective neural architecture search. In Proceedings of the IEEE International Conference on Computer Vision 2019;1304-1313.","DOI":"10.1109\/ICCV.2019.00139"},{"key":"3790_CR31","doi-asserted-by":"crossref","unstructured":"Wen Wei, Liu Hanxiao, Li Hai, Chen Yiran, Bender Gabriel, Kindermans Pieter-Jan. Neural predictor for neural architecture search. arXiv preprint arXiv:1912.00848, 2019.","DOI":"10.1007\/978-3-030-58526-6_39"},{"issue":"35","key":"3790_CR32","first-page":"28037","volume":"6","author":"A Krishnakumar","year":"2022","unstructured":"Krishnakumar A, White C, Zela A, Tu R, Safari M, Hutter F. NAS-Bench-Suite-Zero: accelerating research on zero cost proxies. Adv Neural Inf Proc Syst. 2022;6(35):28037\u201351.","journal-title":"Adv Neural Inf Proc Syst"},{"key":"3790_CR33","unstructured":"Chen Hanlin, Lin Ming, Sun Xiuyu, Li Hao. Nas-bench-zero: A large scale dataset for understanding zero-shot neural architecture search. Openreview preprint https:\/\/openreview.net\/forum?id=hP-SILoczR, 2021."},{"key":"3790_CR34","first-page":"12265","volume":"34","author":"Ning Xuefei","year":"2021","unstructured":"Xuefei Ning, Changcheng Tang, Wenshuo Li, Zixuan Zhou, Shuang Liang, Huazhong Yang, Wang Yu. Evaluating efficient performance estimators of neural architectures. Adv Neural Inf Proc Syst. 2021;34:12265.","journal-title":"Adv Neural Inf Proc Syst"},{"issue":"1","key":"3790_CR35","doi-asserted-by":"publisher","first-page":"72","DOI":"10.2307\/1412159","volume":"15","author":"C Spearman","year":"1904","unstructured":"Spearman C. The proof and measurement of association between two things. Am J Psychol. 1904;15(1):72\u2013101.","journal-title":"Am J Psychol"},{"key":"3790_CR36","unstructured":"Jeevan P, Viswanathan K, Sethi A. Wavemix-lite: A resource-efficient neural network for image analysis. arXiv preprint arXiv:2205.14375. 2022."},{"key":"3790_CR37","unstructured":"Dong Xuanyi, Yang Yi. NAS-Bench-201: Extending the Scope of Reproducible Neural Architecture Search. In ICLR, 2020."},{"key":"3790_CR38","doi-asserted-by":"publisher","unstructured":"Tanveer MS, Karim Khan MU, Kyung C-M. \u201cFine-Tuning DARTS for Image Classification,\u201d 2020 25th International Conference on Pattern Recognition (ICPR), Milan, Italy, 2021;4789-4796, https:\/\/doi.org\/10.1109\/ICPR48806.2021.9412221.","DOI":"10.1109\/ICPR48806.2021.9412221"},{"key":"3790_CR39","unstructured":"Gesmundo A, Dean J. An evolutionary approach to dynamic introduction of tasks in large-scale multitask learning systems. arXiv preprint arXiv:2205.12755. 2022."},{"key":"3790_CR40","unstructured":"DeVries Terrance, Taylor Graham. Improved Regularization of Convolutional Neural Networks with Cutout. 2017."},{"key":"3790_CR41","unstructured":"https:\/\/www.cs.toronto.edu\/~kriz\/cifar.html"},{"key":"3790_CR42","unstructured":"Xiao H, Rasul K, Vollgraf R. Fashion-mnist: a novel image dataset for benchmarking machine learning algorithms. arXiv preprint arXiv:1708.07747. 2017."},{"key":"3790_CR43","doi-asserted-by":"crossref","unstructured":"Cohen G, Afshar S, Tapson J, Van Schaik A. EMNIST: Extending MNIST to handwritten letters. In2017 international joint conference on neural networks (IJCNN) 2017;2921-2926. IEEE.","DOI":"10.1109\/IJCNN.2017.7966217"},{"key":"3790_CR44","unstructured":"Clanuwat T, Bober-Irizar M, Kitamoto A, Lamb A, Yamamoto K, Ha D. Deep learning for classical japanese literature. arXiv preprint arXiv:1812.01718. 2018."},{"key":"3790_CR45","doi-asserted-by":"publisher","first-page":"211","DOI":"10.1007\/s11263-015-0816-y","volume":"115","author":"O Russakovsky","year":"2015","unstructured":"Russakovsky O, Deng J, Su H, Krause J, Satheesh S, Ma S, Huang Z, Karpathy A, Khosla A, Bernstein M, Berg AC. Imagenet large scale visual recognition challenge. Int J Comput vision. 2015;115:211\u201352.","journal-title":"Int J Comput vision"},{"key":"3790_CR46","doi-asserted-by":"crossref","unstructured":"Li G, Qian G, Delgadillo IC, Muller M, Thabet A, Ghanem B. Sgas: Sequential greedy architecture search. InProceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition 2020;1620-1630.","DOI":"10.1109\/CVPR42600.2020.00169"},{"key":"3790_CR47","doi-asserted-by":"crossref","unstructured":"Siddiqui S, Kyrkou C, Theocharides T True Rank Guided Efficient Neural Architecture Search for End to End Low-Complexity Network Discovery. In: Tsapatsoulis, N., et al. Computer Analysis of Images and Patterns. CAIP 2023. 2023.","DOI":"10.1007\/978-3-031-44237-7_3"},{"key":"3790_CR48","unstructured":"Yi D, Lei Z, Liao S, Li SZ. Learning face representation from scratch. arXiv preprint arXiv:1411.7923. (2014)"},{"key":"3790_CR49","unstructured":"Huang GB, Mattar M, Berg T, Learned-Miller E. Labeled faces in the wild: A database forstudying face recognition in unconstrained environments. InWorkshop on faces in\u2019Real-Life\u2019Images: detection, alignment, and recognition 2008."},{"key":"3790_CR50","doi-asserted-by":"crossref","unstructured":"Sengupta S, Chen JC, Castillo C, Patel VM, Chellappa R, Jacobs DW. Frontal to profile face verification in the wild. In2016 IEEE winter conference on applications of computer vision (WACV) 2016;1-9. IEEE.","DOI":"10.1109\/WACV.2016.7477558"},{"issue":"7","key":"3790_CR51","first-page":"5","volume":"5","author":"T Zheng","year":"2018","unstructured":"Zheng T, Deng W. Cross-pose lfw: a database for studying cross-pose face recognition in unconstrained environments. Beijing Univ Posts Telecommun Tech Rep. 2018;5(7):5.","journal-title":"Beijing Univ Posts Telecommun Tech Rep"},{"key":"3790_CR52","first-page":"59","volume":"51","author":"S Moschoglou","year":"2017","unstructured":"Moschoglou S, Papaioannou A, Sagonas C, Deng J, Kotsia I, Zafeiriou S. Agedb: the first manually collected, in-the-wild age database. Inproc IEEE Vonf Comput Vision Pattern Recogn workshops. 2017;51:59.","journal-title":"Inproc IEEE Vonf Comput Vision Pattern Recogn workshops"},{"key":"3790_CR53","unstructured":"Zheng T, Deng W, Hu J. Cross-age lfw: A database for studying cross-age face recognition in unconstrained environments. arXiv preprint arXiv:1708.08197. (2017)."},{"key":"3790_CR54","doi-asserted-by":"crossref","unstructured":"Whitelam C, Taborsky E, Blanton A, Maze B, Adams J, Miller T, Kalka N, Jain AK, Duncan JA, Allen K, Cheney J. Iarpa janus benchmark-b face dataset. Inproceedings of the IEEE conference on computer vision and pattern recognition workshops 2017;90-98.","DOI":"10.1109\/CVPRW.2017.87"},{"key":"3790_CR55","doi-asserted-by":"crossref","unstructured":"Maze B, Adams J, Duncan JA, Kalka N, Miller T, Otto C, Jain AK, Niggel WT, Anderson J, Cheney J, Grother P. Iarpa janus benchmark-c: Face dataset and protocol. In 2018 international conference on biometrics (ICB) 2018;158-165. IEEE.","DOI":"10.1109\/ICB2018.2018.00033"},{"key":"3790_CR56","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-20893-6_38","volume-title":"Low-resolution face recognition","author":"Z Cheng","year":"2019","unstructured":"Cheng Z, Zhu X, Gong S. Low-resolution face recognition. Cham: Springer International Publishing; 2019."},{"key":"3790_CR57","doi-asserted-by":"crossref","unstructured":"Deng J, Guo J, Xue N, Zafeiriou S. Arcface: Additive angular margin loss for deep face recognition. InProceedings of the IEEE\/CVF conference on computer vision and pattern recognition 2019;4690-4699.","DOI":"10.1109\/CVPR.2019.00482"},{"key":"3790_CR58","doi-asserted-by":"crossref","unstructured":"Kim M, Jain AK, Liu X. Adaface: Quality adaptive margin for face recognition. InProceedings of the IEEE\/CVF conference on computer vision and pattern recognition 2022 ;18750-18759.","DOI":"10.1109\/CVPR52688.2022.01819"},{"key":"3790_CR59","first-page":"36","volume":"13","author":"S Dooley","year":"2024","unstructured":"Dooley S, Sukthanker R, Dickerson J, White C, Hutter F, Goldblum M. Rethinking bias mitigation: Fairer architectures make for fairer face recognition. Adv Neural Inf Proc Syst. 2024;13:36.","journal-title":"Adv Neural Inf Proc Syst"},{"issue":"4","key":"3790_CR60","first-page":"2817","volume":"35","author":"X Wang","year":"2021","unstructured":"Wang X. Teacher guided neural architecture search for face recognition. InProc AAAI Conf Artif Intell. 2021;35(4):2817\u201325.","journal-title":"InProc AAAI Conf Artif Intell"},{"issue":"10","key":"3790_CR61","doi-asserted-by":"publisher","first-page":"46823","DOI":"10.1109\/ACCESS.2022.3170561","volume":"26","author":"F Boutros","year":"2022","unstructured":"Boutros F, Siebke P, Klemt M, Damer N, Kirchbuchner F, Kuijper A. Pocketnet: Extreme lightweight face recognition network using neural architecture search and multistep knowledge distillation. IEEE Access. 2022;26(10):46823\u201333.","journal-title":"IEEE Access"}],"container-title":["SN Computer Science"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s42979-025-03790-z.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s42979-025-03790-z\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s42979-025-03790-z.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,3,13]],"date-time":"2025-03-13T05:34:48Z","timestamp":1741844088000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s42979-025-03790-z"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,3,13]]},"references-count":61,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2025,3]]}},"alternative-id":["3790"],"URL":"https:\/\/doi.org\/10.1007\/s42979-025-03790-z","relation":{},"ISSN":["2661-8907"],"issn-type":[{"value":"2661-8907","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,3,13]]},"assertion":[{"value":"4 March 2024","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"8 February 2025","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"13 March 2025","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare that they have no Conflict of interest.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}},{"value":"Not applicable.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Research Involving Humans and\/or Animals"}},{"value":"Informed consent was obtained from all participants included in the study.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Informed Consent"}}],"article-number":"282"}}