{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,3]],"date-time":"2026-05-03T05:30:04Z","timestamp":1777786204489,"version":"3.51.4"},"reference-count":54,"publisher":"Springer Science and Business Media LLC","issue":"19","license":[{"start":{"date-parts":[[2025,5,9]],"date-time":"2025-05-09T00:00:00Z","timestamp":1746748800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,5,9]],"date-time":"2025-05-09T00:00:00Z","timestamp":1746748800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100006704","name":"University of Ostrava","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100006704","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Neural Comput &amp; Applic"],"published-print":{"date-parts":[[2025,7]]},"abstract":"<jats:title>Abstract<\/jats:title>\n          <jats:p>A new concept, termed virtual neural networks, is introduced, where the count of trainable parameters is kept constant, and scalability is attained purely through computational resources. This concept is an abstract framework that can be realized using any standard convolutional neural network. It merges siamese neural networks with a deep ensemble technique by generating numerous virtual models that share weights derived from a small set of physical models. The ensemble comprises up to hundreds of trained models simultaneously. All virtual networks take the same input, and their interconnected structure induces an internal distortion that boosts the entire ensemble robustness. The accuracy of the ensemble improves as the number of virtual networks increases, without changing the capacity. Virtual neural networks outperform larger capacity models, typical deep ensembles, and contemporary approaches like SWA and Masksembles. Additionally, the highest performing individual model from the ensemble surpasses other models trained individually, even those with a greater number of parameters.<\/jats:p>","DOI":"10.1007\/s00521-025-11180-y","type":"journal-article","created":{"date-parts":[[2025,5,9]],"date-time":"2025-05-09T09:53:05Z","timestamp":1746784385000},"page":"14279-14297","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Virtual neural networks: hundreds of souls in a body"],"prefix":"10.1007","volume":"37","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-4349-9705","authenticated-orcid":false,"given":"Petr","family":"Hurtik","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Marek","family":"Vajgl","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zahra","family":"Alijani","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Vojtech","family":"Molek","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"297","published-online":{"date-parts":[[2025,5,9]]},"reference":[{"issue":"3","key":"11180_CR1","doi-asserted-by":"publisher","first-page":"801","DOI":"10.1214\/aos\/1024691079","volume":"26","author":"L Breiman","year":"1998","unstructured":"Breiman L (1998) Arcing classifier (with discussion and a rejoinder by the author). Ann Stat 26(3):801\u2013849","journal-title":"Ann Stat"},{"issue":"2","key":"11180_CR2","doi-asserted-by":"publisher","first-page":"241","DOI":"10.1016\/S0893-6080(05)80023-1","volume":"5","author":"DH Wolpert","year":"1992","unstructured":"Wolpert DH (1992) Stacked generalization. Neural Netw 5(2):241\u2013259","journal-title":"Neural Netw"},{"issue":"2","key":"11180_CR3","doi-asserted-by":"publisher","first-page":"123","DOI":"10.1007\/BF00058655","volume":"24","author":"L Breiman","year":"1996","unstructured":"Breiman L (1996) Bagging predictors. Mach Learn 24(2):123\u2013140","journal-title":"Mach Learn"},{"issue":"4","key":"11180_CR4","doi-asserted-by":"publisher","first-page":"382","DOI":"10.1214\/ss\/1009212519","volume":"14","author":"JA Hoeting","year":"1999","unstructured":"Hoeting JA, Madigan D, Raftery AE, Volinsky CT (1999) Bayesian model averaging: a tutorial (with comments by M. Clyde, David Draper and EJ George, and a rejoinder by the authors. Stat Sci 14(4):382\u2013417","journal-title":"Stat Sci"},{"key":"11180_CR5","doi-asserted-by":"crossref","unstructured":"Monteith K, Carroll James L, Seppi K, Martinez T (2011) Turning Bayesian model averaging into bayesian model combination. In: The 2011 International Joint Conference on Neural Networks, IEEE, p 2657\u20132663.","DOI":"10.1109\/IJCNN.2011.6033566"},{"key":"11180_CR6","unstructured":"Lakshminarayanan B, Pritzel A, Blundell C (2016) Simple and scalable predictive uncertainty estimation using deep ensembles. arXiv preprint arXiv:1612.01474"},{"issue":"4","key":"11180_CR7","doi-asserted-by":"publisher","first-page":"599","DOI":"10.1016\/S0893-6080(96)00098-6","volume":"10","author":"S Hashem","year":"1997","unstructured":"Hashem S (1997) Optimal linear combinations of neural networks. Neural Netw 10(4):599\u2013614","journal-title":"Neural Netw"},{"key":"11180_CR8","unstructured":"Pearlmutter B, Rosenfeld R (1990) Chaitin-kolmogorov complexity and generalization in neural networks. Advances in neural information processing systems, 3"},{"key":"11180_CR9","unstructured":"Buschj\u00e4ger S, Pfahler L, Morik K (2020) Generalized negative correlation learning for deep ensembling. arXiv preprint arXiv:2011.02952"},{"key":"11180_CR10","unstructured":"Krizhevsky A, Sutskever I, Hinton GE (2012) Imagenet classification with deep convolutional neural networks. Advances in neural information processing systems, 25"},{"key":"11180_CR11","unstructured":"Kaggle Team (2017) Carvana image masking challenge-1st place winner\u2019s interview. https:\/\/medium.com\/kaggle-blog\/carvana-image-masking-challenge-1st-place-winners-interview-78fcc5c887a8, Accessed: 2012-02-22"},{"key":"11180_CR12","doi-asserted-by":"crossref","unstructured":"Liu L, Wei W, Chow KH, Loper M, Gursoy E, Truex S, Wu Y (2019) Deep neural network ensembles against deception: Ensemble diversity, accuracy and robustness. In: 2019 IEEE 16th international conference on mobile ad hoc and sensor systems (MASS), IEEE, pages 274\u2013282.","DOI":"10.1109\/MASS.2019.00040"},{"key":"11180_CR13","doi-asserted-by":"crossref","unstructured":"Chicco D (2021) Siamese neural networks: an overview. Artificial Neural Networks, pages 73\u201394,","DOI":"10.1007\/978-1-0716-0826-5_3"},{"key":"11180_CR14","unstructured":"Yang Z, Wang Y, Liu C, Chen H, Xu C, Shi B, Xu C, Xu C (2019) Legonet: Efficient convolutional neural networks with lego filters. In: International Conference on Machine Learning, PMLR, pages 7005\u20137014."},{"key":"11180_CR15","unstructured":"Garipov T, Izmailov P, Podoprikhin D, Vetrov D, Wilson AG (2018) Loss surfaces, mode connectivity, and fast ensembling of dnns. In: Proceedings of the 32nd International Conference on Neural Information Processing Systems, pages 8803\u20138812"},{"key":"11180_CR16","unstructured":"Izmailov P, Podoprikhin D, Garipov T, Vetrov D, Wilson AG (2018) Averaging weights leads to wider optima and better generalization. arXiv preprint arXiv:1803.05407"},{"key":"11180_CR17","doi-asserted-by":"publisher","first-page":"1021","DOI":"10.1109\/LSP.2021.3082036","volume":"28","author":"Y Wang","year":"2021","unstructured":"Wang Y, Park H, Lee J (2021) Memory-free stochastic weight averaging by one-way variational pruning. IEEE Signal Process Lett 28:1021\u20131025","journal-title":"IEEE Signal Process Lett"},{"key":"11180_CR18","unstructured":"Gupta V, Serrano SA, DeCoste D (2020) Stochastic weight averaging in parallel: Large-batch training that generalizes well. arXiv preprint arXiv:2001.02312"},{"key":"11180_CR19","unstructured":"Huang G, Li Y, Pleiss G, Liu Z, Hopcroft JE, Weinberger KQ (2017) Snapshot ensembles: Train 1, get m for free. arXiv preprint arXiv:1704.00109"},{"key":"11180_CR20","unstructured":"Loshchilov I, Hutter F (2016) SGDR: Stochastic gradient descent with warm restarts. arXiv preprint arXiv:1608.03983"},{"key":"11180_CR21","unstructured":"Zhang S, Liu M, Yan J (2020) The diversified ensemble neural network. Advances in Neural Information Processing Systems, 33"},{"issue":"4","key":"11180_CR22","doi-asserted-by":"publisher","first-page":"1090","DOI":"10.1109\/TCSVT.2018.2822773","volume":"29","author":"R Takahashi","year":"2018","unstructured":"Takahashi R, Matsubara T, Uehara K (2018) A novel weight-shared multi-stage CNN for scale robustness. IEEE Trans Circuits Syst Video Technol 29(4):1090\u20131101","journal-title":"IEEE Trans Circuits Syst Video Technol"},{"key":"11180_CR23","unstructured":"Havasi M, Jenatton R, Fort S, Liu JZ, Snoek J, Lakshminarayanan B, Dai AM, Tran D (2020) Training independent subnetworks for robust prediction. arXiv preprint arXiv:2010.06610"},{"issue":"9","key":"11180_CR24","first-page":"1621","volume":"6","author":"G Brown","year":"2005","unstructured":"Brown G, Wyatt JL, Tino P, Bengio Y (2005) Managing diversity in regression ensembles. J Mach Learn Res 6(9):1621\u20131650","journal-title":"J Mach Learn Res"},{"key":"11180_CR25","doi-asserted-by":"crossref","unstructured":"Opitz M, Possegger H, Bischof H (2016) Efficient model averaging for deep neural networks. In: Asian Conference on Computer Vision, Springer, pages 205\u2013220.","DOI":"10.1007\/978-3-319-54184-6_13"},{"key":"11180_CR26","unstructured":"Webb AM, Reynolds C, Iliescu DA, Reeve H, Luj\u00e1n M, Brown G (2019) Joint training of neural network ensembles. stat. 1050:12"},{"key":"11180_CR27","doi-asserted-by":"crossref","unstructured":"Dvornik N, Schmid C, Mairal J (2019) Diversity with cooperation: Ensemble methods for few-shot classification. In: Proceedings of the IEEE\/CVF International Conference on Computer Vision, pages 3723\u20133731","DOI":"10.1109\/ICCV.2019.00382"},{"issue":"1","key":"11180_CR28","first-page":"1929","volume":"15","author":"N Srivastava","year":"2014","unstructured":"Srivastava N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R (2014) Dropout: a simple way to prevent neural networks from overfitting. J Mach Learn Res 15(1):1929\u20131958","journal-title":"J Mach Learn Res"},{"key":"11180_CR29","unstructured":"Baldi P, Sadowski PJ (2013) Understanding dropout. Advances in Neural Information Processing Systems, 26"},{"key":"11180_CR30","unstructured":"Bachman P, Alsharif O, Precup D (2014) Learning with pseudo-ensembles. Advances in Neural Information Processing Systems, 27"},{"key":"11180_CR31","unstructured":"Gal Y, Ghahramani Z (2016) Dropout as a Bayesian approximation: representing model uncertainty in deep learning. In: International Conference on Machine Learning, PMLR, pages 1050\u20131059."},{"key":"11180_CR32","doi-asserted-by":"crossref","unstructured":"Durasov N, Bagautdinov T, Baque P, Fua P (2021) Masksembles for uncertainty estimation. In: Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, pages 13539\u201313548","DOI":"10.1109\/CVPR46437.2021.01333"},{"key":"11180_CR33","doi-asserted-by":"crossref","unstructured":"Kobayashi S, Kiyono S, Suzuki J, Inui K (2022) Diverse lottery tickets boost ensemble from a single pretrained model. In: Challenges & Perspectives in Creating Large Language Models","DOI":"10.18653\/v1\/2022.bigscience-1.4"},{"key":"11180_CR34","unstructured":"Liu S, Chen T, Atashgahi Z, Chen X, Sokar G, Mocanu E, Pechenizkiy M, Wang Z, Mocanu DC (2021) Deep ensembling with no overhead for either training or testing: the all-round blessings of dynamic sparsity. arXiv preprint arXiv:2106.14568"},{"key":"11180_CR35","unstructured":"Zhang Z, Gao VR, Sabuncu MR (2021) Ex uno plures: splitting one model into an ensemble of subnetworks. arXiv preprint arXiv:2106.04767"},{"key":"11180_CR36","doi-asserted-by":"crossref","unstructured":"Bromley J, Guyon I, LeCun Ynn, S\u00e4ckinger E, Shah R (1993) Signature verification using a\" siamese\" time delay neural network. Advances in Neural Information Processing Systems, 6","DOI":"10.1142\/9789812797926_0003"},{"key":"11180_CR37","doi-asserted-by":"crossref","unstructured":"Wang F, Liu H (2021) Understanding the behaviour of contrastive loss. In: Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, pages 2495\u20132504","DOI":"10.1109\/CVPR46437.2021.00252"},{"key":"11180_CR38","doi-asserted-by":"crossref","unstructured":"Taigman Y, Yang M, Ranzato M, Wolf L (2014) Deepface: closing the gap to human-level performance in face verification. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 1701\u20131708","DOI":"10.1109\/CVPR.2014.220"},{"issue":"6","key":"11180_CR39","doi-asserted-by":"publisher","first-page":"10763","DOI":"10.1109\/JIOT.2019.2941527","volume":"6","author":"J Yang","year":"2019","unstructured":"Yang J, Zou H, Zhou Y, Xie L (2019) Learning gestures from wifi: a siamese recurrent convolutional architecture. IEEE Internet Things J 6(6):10763\u201310772","journal-title":"IEEE Internet Things J"},{"key":"11180_CR40","doi-asserted-by":"publisher","first-page":"133924","DOI":"10.1109\/ACCESS.2021.3110297","volume":"9","author":"A Ferreira","year":"2021","unstructured":"Ferreira A, Purnekar N, Barni M (2021) Ensembling shallow siamese neural network architectures for printed documents verification in data-scarcity scenarios. IEEE Access 9:133924\u2013133939","journal-title":"IEEE Access"},{"key":"11180_CR41","doi-asserted-by":"crossref","unstructured":"Huang H, Liu G, Zhang Y, Xiong R, Zhang S (2022) Ensemble siamese networks for object tracking. Neural Computing and Applications, pages 1\u201319","DOI":"10.1007\/s00521-022-06911-4"},{"key":"11180_CR42","unstructured":"Wen Y, Tran D, Ba J (2020) Batchensemble: an alternative approach to efficient ensemble and lifelong learning. arXiv preprint arXiv:2002.06715"},{"key":"11180_CR43","doi-asserted-by":"crossref","unstructured":"Wu Y, Liu L, Xie Z, Chow KH, Wei W (2021) Boosting ensemble accuracy by revisiting ensemble diversity metrics. In: Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, pages 16469\u201316477","DOI":"10.1109\/CVPR46437.2021.01620"},{"key":"11180_CR44","doi-asserted-by":"crossref","unstructured":"Yang Z, Chen W, Wang F, Xu B (2018) Unsupervised neural machine translation with weight sharing. arXiv preprint arXiv:1804.09057","DOI":"10.18653\/v1\/P18-1005"},{"key":"11180_CR45","unstructured":"He F, Liu T, Tao D (2019) Control batch size and learning rate to generalize well: Theoretical and empirical evidence. Advances in Neural Information Processing Systems, 32"},{"key":"11180_CR46","doi-asserted-by":"crossref","unstructured":"Xie S, Girshick R, Doll\u00e1r P, Tu Z, He K (2017) Aggregated residual transformations for deep neural networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 1492\u20131500","DOI":"10.1109\/CVPR.2017.634"},{"key":"11180_CR47","doi-asserted-by":"crossref","unstructured":"He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 770\u2013778","DOI":"10.1109\/CVPR.2016.90"},{"key":"11180_CR48","unstructured":"Tan M, Le Q (2019) Efficientnet: rethinking model scaling for convolutional neural networks. In: International Conference on Machine Learning, PMLR, pages 6105\u20136114"},{"key":"11180_CR49","volume-title":"Learning multiple layers of features from tiny images","author":"A Krizhevsky","year":"2009","unstructured":"Krizhevsky A, Hinton G (2009) Learning multiple layers of features from tiny images. Technical report, Citeseer"},{"key":"11180_CR50","unstructured":"Cao K, Wei C, Gaidon A, Arechiga N, Ma T (2019) Learning imbalanced datasets with label-distribution-aware margin loss. In: Advances in Neural Information Processing Systems"},{"issue":"3","key":"11180_CR51","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, Hao S, Krause J, Satheesh S, Ma S, Huang Z, Karpathy A, Khosla A, Bernstein M, Berg AC, Fei-Fei L (2015) ImageNet large scale visual recognition challenge. Int J Comput Vision (IJCV) 115(3):211\u2013252","journal-title":"Int J Comput Vision (IJCV)"},{"key":"11180_CR52","unstructured":"Kingma DP, Ba J (2014) Adam: a method for stochastic optimization. arXiv preprint arXiv:1412.6980"},{"key":"11180_CR53","doi-asserted-by":"crossref","unstructured":"Akiba T, Sano S, Yanase T, Ohta T, Koyama M (2019) Optuna: a next-generation hyperparameter optimization framework. In: Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pages 2623\u20132631","DOI":"10.1145\/3292500.3330701"},{"key":"11180_CR54","unstructured":"Dehghani M, Arnab A, Beyer L, Vaswani A, Tay Y (2021) The efficiency misnomer. arXiv preprint arXiv:2110.12894"}],"container-title":["Neural Computing and Applications"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00521-025-11180-y.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s00521-025-11180-y\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00521-025-11180-y.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,27]],"date-time":"2025-06-27T16:02:52Z","timestamp":1751040172000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s00521-025-11180-y"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,5,9]]},"references-count":54,"journal-issue":{"issue":"19","published-print":{"date-parts":[[2025,7]]}},"alternative-id":["11180"],"URL":"https:\/\/doi.org\/10.1007\/s00521-025-11180-y","relation":{},"ISSN":["0941-0643","1433-3058"],"issn-type":[{"value":"0941-0643","type":"print"},{"value":"1433-3058","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,5,9]]},"assertion":[{"value":"9 May 2024","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"7 March 2025","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"9 May 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":"This article does not contain any studies with human participants or animals performed by any of the authors.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Human or animal consent"}}]}}