{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,13]],"date-time":"2025-10-13T04:15:46Z","timestamp":1760328946109,"version":"3.37.3"},"reference-count":51,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","license":[{"start":{"date-parts":[[2024,1,1]],"date-time":"2024-01-01T00:00:00Z","timestamp":1704067200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Access"],"published-print":{"date-parts":[[2024]]},"DOI":"10.1109\/access.2024.3486085","type":"journal-article","created":{"date-parts":[[2024,10,24]],"date-time":"2024-10-24T17:28:48Z","timestamp":1729790928000},"page":"161312-161328","source":"Crossref","is-referenced-by-count":1,"title":["Faster Training of Large Kernel Convolutions on Smaller Spatial Scales"],"prefix":"10.1109","volume":"12","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-3527-690X","authenticated-orcid":false,"given":"Shota","family":"Fukuzaki","sequence":"first","affiliation":[{"name":"Department of Electronics and Electrical Engineering, Keio University, Yokohama-shi, Kanagawa, Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3461-1507","authenticated-orcid":false,"given":"Masaaki","family":"Ikehara","sequence":"additional","affiliation":[{"name":"Department of Electronics and Electrical Engineering, Keio University, Yokohama-shi, Kanagawa, Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","article-title":"An image is worth 16\u00d716 words: Transformers for image recognition at scale","author":"Dosovitskiy","year":"2020","journal-title":"arXiv:2010.11929"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00061"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00986"},{"key":"ref4","first-page":"3965","article-title":"CoAtNet: Marrying convolution and attention for all data sizes","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"34","author":"Dai"},{"key":"ref5","first-page":"23495","article-title":"Inception transformer","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"35","author":"Si"},{"key":"ref6","doi-asserted-by":"crossref","DOI":"10.1109\/TPAMI.2023.3341806","article-title":"CrossFormer: A versatile vision transformer hinging on cross-scale attention","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Wang"},{"key":"ref7","article-title":"Very deep convolutional networks for large-scale image recognition","author":"Simonyan","year":"2014","journal-title":"arXiv:1409.1556"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01166"},{"key":"ref9","first-page":"1","article-title":"PyTorch: An imperative style, high-performance deep learning library","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"32","author":"Paszke"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-015-0816-y"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/GCCE59613.2023.10315578"},{"article-title":"Learning multiple layers of features from tiny images","year":"2009","author":"Krizhevsky","key":"ref12"},{"key":"ref13","first-page":"215","article-title":"An analysis of single layer networks in unsupervised feature learning","volume-title":"Proc. AISTATS","author":"Coates"},{"key":"ref14","first-page":"249","article-title":"Understanding the difficulty of training deep feedforward neural networks","volume-title":"Proc. 13th Int. Conf. Artif. Intell. Statist.","author":"Glorot"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.123"},{"key":"ref16","article-title":"Adam: A method for stochastic optimization","author":"Kingma","year":"2014","journal-title":"arXiv:1412.6980"},{"article-title":"Decoupled weight decay regularization","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Loshchilov","key":"ref17"},{"article-title":"Sharpness-aware minimization for efficiently improving generalization","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Foret","key":"ref18"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2014.222"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00502"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/WACV.2017.58"},{"key":"ref22","first-page":"369","article-title":"Super-convergence: Very fast training of neural networks using large learning rates","volume-title":"Proc. SPIE","volume":"11006","author":"Smith"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1503.02531"},{"key":"ref24","first-page":"10347","article-title":"Training data-efficient image transformers & distillation through attention","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Touvron"},{"article-title":"mixup: Beyond empirical risk minimization","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Zhang","key":"ref25"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW50498.2020.00359"},{"issue":"1","key":"ref27","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":"ref28","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46493-0_39"},{"key":"ref29","first-page":"10096","article-title":"EfficientNetv2: Smaller models and faster training","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Tan"},{"key":"ref30","first-page":"6105","article-title":"EfficientNet: Rethinking model scaling for convolutional neural networks","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Tan"},{"key":"ref31","article-title":"Scale-invariant convolutional neural networks","author":"Xu","year":"2014","journal-title":"arXiv:1411.6369"},{"key":"ref32","article-title":"Locally scale-invariant convolutional neural networks","author":"Kanazawa","year":"2014","journal-title":"arXiv:1412.5104"},{"key":"ref33","article-title":"Scale steerable filters for locally scale-invariant convolutional neural networks","author":"Ghosh","year":"2019","journal-title":"arXiv:1906.03861"},{"article-title":"Scale-equivariant steerable networks","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Sosnovik","key":"ref34"},{"key":"ref35","article-title":"ImageNet classification with deep convolutional neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"25","author":"Krizhevsky"},{"key":"ref36","first-page":"4278","article-title":"Inception-v4, inception-ResNet and the impact of residual connections on learning","volume-title":"Proc. AAAI Conf. Artif. Intell.","volume":"31","author":"Szegedy"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.189"},{"key":"ref38","article-title":"Attention is all you need","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"30","author":"Vaswani"},{"key":"ref39","first-page":"24261","article-title":"MLP-mixer: An all-MLP architecture for vision","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"34","author":"Tolstikhin"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01167"},{"article-title":"More ConvNets in the 2020s: Scaling up kernels beyond 51\u00d751 using sparsity","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Liu","key":"ref41"},{"article-title":"Conditional positional encodings for vision transformers","volume-title":"Proc. 11th Int. Conf. Learn. Represent.","author":"Chu","key":"ref42"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.5555\/3045118.3045167"},{"article-title":"Characterizing signal propagation to close the performance gap in unnormalized resnets","volume-title":"Proc. 9th Int. Conf. Learn. Represent.","author":"Brock","key":"ref44"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref46","article-title":"Micro-batch training with batch-channel normalization and weight standardization","author":"Qiao","year":"2019","journal-title":"arXiv:1903.10520"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01352"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.308"},{"key":"ref49","first-page":"1059","article-title":"High-performance large-scale image recognition without normalization","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Brock"},{"volume-title":"JAX: Composable Transformations of Python+NumPy Programs","year":"2018","author":"Bradbury","key":"ref50"},{"key":"ref51","article-title":"A study of BFLOAT16 for deep learning training","author":"Kalamkar","year":"2019","journal-title":"arXiv:1905.12322"}],"container-title":["IEEE Access"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/6287639\/10380310\/10734125.pdf?arnumber=10734125","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,11,27]],"date-time":"2024-11-27T02:35:49Z","timestamp":1732674949000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10734125\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024]]},"references-count":51,"URL":"https:\/\/doi.org\/10.1109\/access.2024.3486085","relation":{},"ISSN":["2169-3536"],"issn-type":[{"type":"electronic","value":"2169-3536"}],"subject":[],"published":{"date-parts":[[2024]]}}}