{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,20]],"date-time":"2026-03-20T00:39:53Z","timestamp":1773967193950,"version":"3.50.1"},"reference-count":54,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2026,3,1]],"date-time":"2026-03-01T00:00:00Z","timestamp":1772323200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/tdm\/userlicense\/1.0\/"},{"start":{"date-parts":[[2026,3,1]],"date-time":"2026-03-01T00:00:00Z","timestamp":1772323200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/legal\/tdmrep-license"},{"start":{"date-parts":[[2026,3,1]],"date-time":"2026-03-01T00:00:00Z","timestamp":1772323200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-017"},{"start":{"date-parts":[[2026,3,1]],"date-time":"2026-03-01T00:00:00Z","timestamp":1772323200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"},{"start":{"date-parts":[[2026,3,1]],"date-time":"2026-03-01T00:00:00Z","timestamp":1772323200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-012"},{"start":{"date-parts":[[2026,3,1]],"date-time":"2026-03-01T00:00:00Z","timestamp":1772323200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,3,1]],"date-time":"2026-03-01T00:00:00Z","timestamp":1772323200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-004"}],"funder":[{"DOI":"10.13039\/100004358","name":"Samsung","doi-asserted-by":"publisher","award":["IO201210-08019-01"],"award-info":[{"award-number":["IO201210-08019-01"]}],"id":[{"id":"10.13039\/100004358","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["elsevier.com","sciencedirect.com"],"crossmark-restriction":true},"short-container-title":["Applied Soft Computing"],"published-print":{"date-parts":[[2026,3]]},"DOI":"10.1016\/j.asoc.2026.114611","type":"journal-article","created":{"date-parts":[[2026,1,12]],"date-time":"2026-01-12T07:33:27Z","timestamp":1768203207000},"page":"114611","update-policy":"https:\/\/doi.org\/10.1016\/elsevier_cm_policy","source":"Crossref","is-referenced-by-count":0,"special_numbering":"C","title":["Stochastic subsampling with average pooling"],"prefix":"10.1016","volume":"190","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-4155-9225","authenticated-orcid":false,"given":"Bum Jun","family":"Kim","sequence":"first","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6023-1837","authenticated-orcid":false,"given":"Sang Woo","family":"Kim","sequence":"additional","affiliation":[]}],"member":"78","reference":[{"key":"10.1016\/j.asoc.2026.114611_bib0005","doi-asserted-by":"crossref","DOI":"10.1016\/j.asoc.2024.112165","article-title":"Multi-fidelity deep neural network with monte carlo dropout technique for uncertainty-aware risk recognition of backward erosion piping in dikes","volume":"166","author":"Liu","year":"2024","journal-title":"Appl. Soft Comput."},{"key":"10.1016\/j.asoc.2026.114611_bib0010","doi-asserted-by":"crossref","DOI":"10.1016\/j.asoc.2020.106209","article-title":"Mixed spatial pyramid pooling for semantic segmentation","volume":"91","author":"Xia","year":"2020","journal-title":"Appl. Soft Comput."},{"key":"10.1016\/j.asoc.2026.114611_bib0015","doi-asserted-by":"crossref","DOI":"10.1016\/j.asoc.2024.111337","article-title":"Regularizing deep neural networks for medical image analysis with augmented batch normalization","volume":"154","author":"Zhu","year":"2024","journal-title":"Appl. Soft Comput."},{"key":"10.1016\/j.asoc.2026.114611_bib0020","doi-asserted-by":"crossref","DOI":"10.1016\/j.asoc.2024.111705","article-title":"Sparsify dynamically expandable network via variational dropout","volume":"160","author":"Yang","year":"2024","journal-title":"Appl. Soft Comput."},{"key":"10.1016\/j.asoc.2026.114611_bib0025","article-title":"An optimized convolutional neural network with a novel spherical triangular fuzzy pooling layer for an algorithmic trading model","volume":"182","author":"Amiri","year":"2025","journal-title":"Appl. Soft Comput."},{"key":"10.1016\/j.asoc.2026.114611_bib0030","series-title":"ICLR","article-title":"Three mechanisms of weight decay regularization","author":"Zhang","year":"2019"},{"key":"10.1016\/j.asoc.2026.114611_bib0035","series-title":"ICML","first-page":"448","article-title":"Batch normalization: accelerating deep network training by reducing internal covariate shift","volume":"vol. 37","author":"Ioffe","year":"2015"},{"key":"10.1016\/j.asoc.2026.114611_bib0040","first-page":"1929","article-title":"Dropout: a simple way to prevent neural networks from overfitting","volume":"15","author":"Srivastava","year":"2014","journal-title":"J. Mach. Learn. Res."},{"key":"10.1016\/j.asoc.2026.114611_bib0045","series-title":"CVPR","first-page":"2682","article-title":"Understanding the disharmony between dropout and batch normalization by variance shift","author":"Li","year":"2019"},{"key":"10.1016\/j.asoc.2026.114611_bib0050","series-title":"UAI","first-page":"1058","article-title":"How to use dropout correctly on residual networks with batch normalization","volume":"vol. 216","author":"Kim","year":"2023"},{"key":"10.1016\/j.asoc.2026.114611_bib0055","series-title":"WACV","first-page":"3942","article-title":"PatchDropout: economizing vision transformers using patch dropout","author":"Liu","year":"2023"},{"key":"10.1016\/j.asoc.2026.114611_bib0060","series-title":"CVPR","first-page":"15979","article-title":"Masked autoencoders are scalable vision learners","author":"He","year":"2022"},{"key":"10.1016\/j.asoc.2026.114611_bib0065","series-title":"CVPR","first-page":"23390","article-title":"Scaling Language-Image Pre-Training via masking","author":"Li","year":"2023"},{"key":"10.1016\/j.asoc.2026.114611_bib0070","series-title":"ICLR","article-title":"Stochastic pooling for regularization of deep convolutional neural networks","author":"Zeiler","year":"2013"},{"key":"10.1016\/j.asoc.2026.114611_bib0075","series-title":"ICML","first-page":"43205","article-title":"Provable multi-instance deep AUC maximization with stochastic pooling","volume":"vol. 202","author":"Zhu","year":"2023"},{"key":"10.1016\/j.asoc.2026.114611_bib0080","series-title":"NIPS","first-page":"971","article-title":"Self-Normalizing neural networks","author":"Klambauer","year":"2017"},{"key":"10.1016\/j.asoc.2026.114611_bib0085","series-title":"ECCV","first-page":"646","article-title":"Deep networks with stochastic depth","volume":"vol. 9908","author":"Huang","year":"2016"},{"key":"10.1016\/j.asoc.2026.114611_bib0090","series-title":"ICLR","article-title":"Token merging: your ViT but faster","author":"Bolya","year":"2023"},{"key":"10.1016\/j.asoc.2026.114611_bib0095","series-title":"WACV","first-page":"86","article-title":"GTP-ViT: efficient vision transformers via graph-based token propagation","author":"Xu","year":"2024"},{"key":"10.1016\/j.asoc.2026.114611_bib0100","series-title":"CVPR","first-page":"16070","article-title":"Zero-TPrune: Zero-Shot token pruning through leveraging of the attention graph in Pre-Trained transformers","author":"Wang","year":"2024"},{"key":"10.1016\/j.asoc.2026.114611_bib0105","series-title":"ECCV","first-page":"325","article-title":"Removing rows and columns of tokens in vision transformer enables faster dense prediction without retraining","volume":"vol. 15130","author":"Su","year":"2024"},{"key":"10.1016\/j.asoc.2026.114611_bib0110","series-title":"CVPR","first-page":"2092","article-title":"Joint token pruning and squeezing towards more aggressive compression of vision transformers","author":"Wei","year":"2023"},{"key":"10.1016\/j.asoc.2026.114611_bib0115","doi-asserted-by":"crossref","DOI":"10.1016\/j.neunet.2024.106340","article-title":"Life regression based patch slimming for vision transformers","volume":"176","author":"Chen","year":"2024","journal-title":"Neural Networks"},{"key":"10.1016\/j.asoc.2026.114611_bib0120","doi-asserted-by":"crossref","DOI":"10.1016\/j.imavis.2024.105239","article-title":"Simultaneous image patch attention and pruning for patch selective transformer","volume":"150","author":"Kim","year":"2024","journal-title":"Image Vis. Comput."},{"key":"10.1016\/j.asoc.2026.114611_bib0125","doi-asserted-by":"crossref","DOI":"10.1016\/j.eswa.2025.127449","article-title":"Efficient token pruning in vision transformers using an attention-based multilayer network","volume":"279","author":"Marchetti","year":"2025","journal-title":"Expert Syst. Appl."},{"key":"10.1016\/j.asoc.2026.114611_bib0130","series-title":"ICLR","article-title":"Very deep convolutional networks for Large-Scale image recognition","author":"Simonyan","year":"2015"},{"key":"10.1016\/j.asoc.2026.114611_bib0135","series-title":"ICML","first-page":"6105","article-title":"EfficientNet: rethinking model scaling for convolutional neural networks","volume":"vol. 97","author":"Tan","year":"2019"},{"key":"10.1016\/j.asoc.2026.114611_bib0140","series-title":"ICML","first-page":"1059","article-title":"High-Performance Large-Scale image recognition without normalization","volume":"vol. 139","author":"Brock","year":"2021"},{"key":"10.1016\/j.asoc.2026.114611_bib0145","series-title":"ICLR","article-title":"An image is worth 16x16 words: transformers for image recognition at scale","author":"Dosovitskiy","year":"2021"},{"key":"10.1016\/j.asoc.2026.114611_bib0150","series-title":"CVPR","first-page":"770","article-title":"Deep residual learning for image recognition","author":"He","year":"2016"},{"key":"10.1016\/j.asoc.2026.114611_bib0155","series-title":"ICLR","article-title":"Network in network","author":"Lin","year":"2014"},{"key":"10.1016\/j.asoc.2026.114611_bib0160","series-title":"CVPR","first-page":"7132","article-title":"Squeeze-and-Excitation networks","author":"Hu","year":"2018"},{"key":"10.1016\/j.asoc.2026.114611_bib0165","series-title":"CVPR","first-page":"510","article-title":"Selective kernel networks","author":"Li","year":"2019"},{"key":"10.1016\/j.asoc.2026.114611_bib0170","series-title":"CVPR","first-page":"11531","article-title":"ECA-net: efficient channel attention for deep convolutional neural networks","author":"Wang","year":"2020"},{"key":"10.1016\/j.asoc.2026.114611_bib0175","series-title":"CVPR","first-page":"6230","article-title":"Pyramid scene parsing network","author":"Zhao","year":"2017"},{"key":"10.1016\/j.asoc.2026.114611_bib0180","series-title":"ECCV","first-page":"432","article-title":"Unified perceptual parsing for scene understanding","volume":"vol. 11209","author":"Xiao","year":"2018"},{"key":"10.1016\/j.asoc.2026.114611_bib0185","series-title":"CVPR","first-page":"558","article-title":"Bag of tricks for image classification with convolutional neural networks","author":"He","year":"2019"},{"key":"10.1016\/j.asoc.2026.114611_bib0190","series-title":"ECCV","first-page":"459","article-title":"MaxViT: multi-axis vision transformer","volume":"vol. 13684","author":"Tu","year":"2022"},{"key":"10.1016\/j.asoc.2026.114611_bib0195","series-title":"NeurIPS","first-page":"8024","article-title":"PyTorch: an imperative style, High-Performance deep learning library","author":"Paszke","year":"2019"},{"key":"10.1016\/j.asoc.2026.114611_bib0200","series-title":"WACV","first-page":"1399","article-title":"TResNet: high performance GPU-dedicated architecture","author":"Ridnik","year":"2021"},{"key":"10.1016\/j.asoc.2026.114611_bib0205","series-title":"NeurIPS","first-page":"10750","article-title":"DropBlock: a regularization method for convolutional networks","author":"Ghiasi","year":"2018"},{"key":"10.1016\/j.asoc.2026.114611_bib0210","series-title":"AAAI","first-page":"9351","article-title":"AutoDropout: learning dropout patterns to regularize deep networks","author":"Pham","year":"2021"},{"key":"10.1016\/j.asoc.2026.114611_bib0215","unstructured":"A. Krizhevsky, G. Hinton, et al., Learning multiple layers of features from tiny images, 2009, Toronto, ON, Canada."},{"key":"10.1016\/j.asoc.2026.114611_bib0220","series-title":"CVPR","first-page":"3498","article-title":"Cats and dogs","author":"Parkhi","year":"2012"},{"key":"10.1016\/j.asoc.2026.114611_bib0225","doi-asserted-by":"crossref","first-page":"59","DOI":"10.1016\/j.cviu.2005.09.012","article-title":"Learning generative visual models from few training examples: an incremental Bayesian approach tested on 101 object categories","volume":"106","author":"Fei-Fei","year":"2007","journal-title":"Comput. Vis. Image Underst."},{"key":"10.1016\/j.asoc.2026.114611_bib0230","series-title":"ICCV Workshops","first-page":"554","article-title":"3d object representations for Fine-Grained categorization","author":"Krause","year":"2013"},{"key":"10.1016\/j.asoc.2026.114611_bib0235","first-page":"293","article-title":"The ISPRS benchmark on urban object classification and 3d building reconstruction","volume":"1","author":"Rottensteiner","year":"2012","journal-title":"ISPRS Ann. Photogramm. Remote Sens. Spat. Inf. Sci."},{"key":"10.1016\/j.asoc.2026.114611_bib0240","series-title":"ECCV","first-page":"740","article-title":"Microsoft COCO: common objects in context","volume":"vol. 8693","author":"Lin","year":"2014"},{"key":"10.1016\/j.asoc.2026.114611_bib0245","series-title":"NIPS","first-page":"2814","article-title":"Understanding dropout","author":"Baldi","year":"2013"},{"key":"10.1016\/j.asoc.2026.114611_bib0250","series-title":"CVPR","first-page":"248","article-title":"ImageNet: a large-scale hierarchical image database","author":"Deng","year":"2009"},{"key":"10.1016\/j.asoc.2026.114611_bib0255","series-title":"ICLR","article-title":"SGDR: stochastic gradient descent with warm restarts","author":"Loshchilov","year":"2017"},{"key":"10.1016\/j.asoc.2026.114611_bib0260","author":"Contributors"},{"key":"10.1016\/j.asoc.2026.114611_bib0265","series-title":"CVPR","first-page":"7151","article-title":"Context encoding for semantic segmentation","author":"Zhang","year":"2018"},{"key":"10.1016\/j.asoc.2026.114611_bib0270","series-title":"CVPR","first-page":"7373","article-title":"Dynamic head: unifying object detection heads with attentions","author":"Dai","year":"2021"}],"container-title":["Applied Soft Computing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S1568494626000591?httpAccept=text\/xml","content-type":"text\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S1568494626000591?httpAccept=text\/plain","content-type":"text\/plain","content-version":"vor","intended-application":"text-mining"}],"deposited":{"date-parts":[[2026,3,19]],"date-time":"2026-03-19T22:43:36Z","timestamp":1773960216000},"score":1,"resource":{"primary":{"URL":"https:\/\/linkinghub.elsevier.com\/retrieve\/pii\/S1568494626000591"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3]]},"references-count":54,"alternative-id":["S1568494626000591"],"URL":"https:\/\/doi.org\/10.1016\/j.asoc.2026.114611","relation":{},"ISSN":["1568-4946"],"issn-type":[{"value":"1568-4946","type":"print"}],"subject":[],"published":{"date-parts":[[2026,3]]},"assertion":[{"value":"Elsevier","name":"publisher","label":"This article is maintained by"},{"value":"Stochastic subsampling with average pooling","name":"articletitle","label":"Article Title"},{"value":"Applied Soft Computing","name":"journaltitle","label":"Journal Title"},{"value":"https:\/\/doi.org\/10.1016\/j.asoc.2026.114611","name":"articlelink","label":"CrossRef DOI link to publisher maintained version"},{"value":"article","name":"content_type","label":"Content Type"},{"value":"\u00a9 2026 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.","name":"copyright","label":"Copyright"}],"article-number":"114611"}}