{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,25]],"date-time":"2026-03-25T15:45:48Z","timestamp":1774453548932,"version":"3.50.1"},"reference-count":39,"publisher":"Tech Science Press","issue":"2","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["CMC"],"published-print":{"date-parts":[[2025]]},"DOI":"10.32604\/cmc.2025.063723","type":"journal-article","created":{"date-parts":[[2025,6,24]],"date-time":"2025-06-24T03:28:02Z","timestamp":1750735682000},"page":"3241-3257","source":"Crossref","is-referenced-by-count":2,"title":["Low-Complexity Hardware Architecture for Batch Normalization of CNN Training Accelerator"],"prefix":"10.32604","volume":"84","author":[{"given":"Go-Eun","family":"Woo","sequence":"first","affiliation":[]},{"given":"Sang-Bo","family":"Park","sequence":"additional","affiliation":[]},{"given":"Gi-Tae","family":"Park","sequence":"additional","affiliation":[]},{"given":"Muhammad","family":"Junaid","sequence":"additional","affiliation":[]},{"given":"Hyung-Won","family":"Kim","sequence":"additional","affiliation":[]}],"member":"17807","published-online":{"date-parts":[[2025]]},"reference":[{"key":"ref1","doi-asserted-by":"crossref","first-page":"436","DOI":"10.1038\/nature14539","article-title":"Deep learning","volume":"521","author":"LeCun","year":"2015","journal-title":"Nature"},{"key":"ref2","doi-asserted-by":"crossref","first-page":"107","DOI":"10.1142\/S0218488598000094","article-title":"The vanishing gradient problem during learning recurrent neural nets and problem solutions","volume":"6","author":"Hochreiter","year":"1998","journal-title":"Int J Uncertain Fuzz Knowl Based Syst"},{"key":"ref3","volume":"31","author":"Hanin","year":"2018","journal-title":"Advances in neural information processing systems"},{"key":"ref4","doi-asserted-by":"crossref","first-page":"022022","DOI":"10.1088\/1742-6596\/1168\/2\/022022","article-title":"An overview of overfitting and its solutions","volume":"1168","author":"Ying","year":"2019","journal-title":"J Phy Conf Series"},{"key":"ref5","series-title":"Proceedings of the 27th International Conference on Machine Learning (ICML-10)","first-page":"807","article-title":"Rectified linear units improve restricted Boltzmann machines","author":"Nair","year":"2010 Jun 21\u201324"},{"key":"ref6","series-title":"International Conference on Learning Representations ICLR 2020","article-title":"Why gradient clipping accelerates training: a theoretical justification for adaptivity","author":"Zhang","year":"2020 Apr 30"},{"key":"ref7","first-page":"1929","article-title":"Dropout: a simple way to prevent neural networks from overfitting","volume":"15","author":"Nitish","year":"2014","journal-title":"J Mach Learn Res"},{"key":"ref8","unstructured":"Ioffe S, Szegedy C. Batch normalization: accelerating deep network training by reducing internal covariate shift. arXiv:1502.03167. 2015."},{"key":"ref9","series-title":"Lake Tahoe, NV, USA","volume":"25","author":"Krizhevsky","year":"2012","journal-title":"Advances in neural information processing systems"},{"key":"ref10","volume":"31","author":"Santurkar","year":"2018","journal-title":"Advances in neural information processing systems"},{"key":"ref11","unstructured":"Li Y, Wang N, Shi J, Liu J, Hou X. Revisiting batch normalization for practical domain adaptation. arXiv:1603.04779. 2016."},{"key":"ref12","unstructured":"Summers C, Dinneen MJ. Four things everyone should know to improve batch normalization. arXiv:1906.03548. 2019."},{"key":"ref13","volume":"31","author":"Bjorck","year":"2018","journal-title":"Advances in neural information processing systems"},{"key":"ref14","first-page":"5875","article-title":"New interpretations of normalization methods in deep learning","volume":"34","author":"Sun","year":"2020","journal-title":"Proc AAAI Conf Artif Intell"},{"key":"ref15","volume":"30","author":"Ioffe","year":"2017","journal-title":"Advances in neural information processing systems"},{"key":"ref16","doi-asserted-by":"crossref","first-page":"12777","DOI":"10.1007\/s11042-019-08453-9","article-title":"Dropout vs. batch normalization: an empirical study of their impact to deep learning","volume":"79","author":"Garbin","year":"2020","journal-title":"Multimed Tools Appl"},{"key":"ref17","series-title":"2018 Fifth International Conference on Emerging Applications of Information Technology (EAIT); 2018 Jan 12\u201313; Kolkata, India","first-page":"1","article-title":"Batch normalization in convolutional neural networks\u2014a comparative study with CIFAR-10 data","author":"Thakkar"},{"key":"ref18","first-page":"4778","article-title":"Beyond batchnorm: towards a unified understanding of normalization in deep learning","volume":"34","author":"Lubana","year":"2021","journal-title":"Adv Neural Inf Process Syst"},{"key":"ref19","doi-asserted-by":"crossref","first-page":"2145","DOI":"10.3390\/s24072145","article-title":"Hybrid Precision Floating-Point (HPFP) selection to optimize hardware-constrained accelerator for CNN training","volume":"24","author":"Junaid","year":"2024","journal-title":"Sensors"},{"key":"ref20","unstructured":"Ba JL, Kiros JR, Hinton GE. Layer normalization. arXiv:1607.06450. 2016."},{"key":"ref21","series-title":"Proceedings of the 2017 IEEE International Conference on Computer Vision (ICCV)","first-page":"1501","article-title":"Arbitrary style transfer in real-time with adaptive instance normalization","author":"Huang","year":"2017 Oct 22\u201329"},{"key":"ref22","series-title":"Proceedings of the 2018 European Conference on Computer Vision (ECCV)","first-page":"3","article-title":"Group normalization","author":"Wu","year":"2018 Sep 8\u201314"},{"key":"ref23","doi-asserted-by":"crossref","first-page":"533","DOI":"10.1038\/323533a0","article-title":"Learning representations by back-propagating errors","volume":"323","author":"Rumelhart","year":"1986","journal-title":"Nature"},{"key":"ref24","volume":"31","author":"Banner","year":"2018","journal-title":"Advances in neural information processing systems"},{"key":"ref25","volume":"27","author":"Soudry","year":"2014","journal-title":"Advances in neural information processing systems"},{"key":"ref26","doi-asserted-by":"crossref","first-page":"29","DOI":"10.1109\/LES.2020.2975055","article-title":"Bactran: a hardware batch normalization implementation for CNN training engine","volume":"13","author":"Yang","year":"2021","journal-title":"IEEE Embedd Syst Lett"},{"key":"ref27","series-title":"2022 IEEE 40th International Conference on Computer Design (ICCD)","first-page":"443","article-title":"LightNorm: Area and Energy-Efficient Batch Normalization Hardware for On-Device DNN Training","author":"Noh","year":"2022 Oct 23\u201326"},{"key":"ref28","doi-asserted-by":"crossref","first-page":"738","DOI":"10.1109\/TVLSI.2023.3262787","article-title":"ACBN: approximate calculated batch normalization for efficient DNN on-device training processor","volume":"31","author":"Li","year":"2023","journal-title":"IEEE Trans Very Large Scale Integr (VLSI) Syst"},{"key":"ref29","doi-asserted-by":"crossref","first-page":"2043","DOI":"10.1109\/TNNLS.2018.2876179","article-title":"L1-Norm batch normalization for efficient training of deep neural networks","volume":"30","author":"Wu","year":"Jul 2019","journal-title":"IEEE Trans Neural Netw Learn Syst"},{"key":"ref30","series-title":"2021 IEEE International Symposium on Circuits and Systems (ISCAS)","first-page":"1","article-title":"Batch normalization processor design for convolution neural network training and inference","author":"Ting","year":"2021 May 22\u201328"},{"key":"ref31","series-title":"2019 IEEE 13th International Conference on ASIC (ASICON); 2019 Oct 29\u2013Nov 1; Chongqing, China","first-page":"1","article-title":"BNReLU: combine batch normalization and rectified linear unit to reduce hardware overhead","author":"Ge"},{"key":"ref32","series-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition; 2020 Jun 13\u201319; Seattle, WA, USA","first-page":"11237","article-title":"Filter response normalization layer: Eliminating batch dependence in the training of deep neural networks","author":"Singh"},{"key":"ref33","series-title":"2023 IEEE International Symposium on Circuits and Systems (ISCAS); 2023 May 21\u201325; Monterey, CA, USA","first-page":"1","article-title":"Low-cost high-precision architecture for arbitrary floating-point nth root computation","author":"Hong","year":"2023"},{"key":"ref34","series-title":"Proceedings of the 1998 ACM\/SIGDA Sixth International Symposium on Field Programmable Gate Arrays; 1998 Feb 22\u201325; Monterey, CA, USA","first-page":"191","article-title":"A survey of CORDIC algorithms for FPGA based computers","author":"Ray"},{"key":"ref35","series-title":"2005 IEEE International Conference on Application-Specific Systems, Architecture Processors (ASAP\u201905); 2005 Jul 23\u201325; Samos, Greece","first-page":"309","article-title":"Decimal floating-point square root using Newton-Raphson iteration","author":"Wang"},{"key":"ref36","series-title":"2021 IEEE International Midwest Symposium on Circuits and Systems (MWSCAS); 2021 Aug 9\u201311; Lansing, MI, USA","first-page":"774","article-title":"Floating-point square root calculation algorithm based on taylor-series expansion and region division","author":"Wei"},{"key":"ref37","series-title":"51st Midwest Symposium on Circuits and Systems","first-page":"954","article-title":"Floating-point division and square root implementation using a taylor-series expansion algorithm with reduced look-up tables","author":"Kwon","year":"2008 Aug 10\u201313"},{"key":"ref38","first-page":"19","article-title":"Cubic spline interpolation square-root compute unit for cost-efficient batch-normalization calculation of accurate DNN training","volume":"62","author":"Woo","year":"2025","journal-title":"J Inst Electron Inf Eng"},{"key":"ref39","unstructured":"Park SB. Reconfigurable CNN Training Accelerator Design Based on Efficient Memory Access Reduction Techniques [master\u2019s thesis]. Cheongju, Republic of Korea: Chungbuk National University; 2024."}],"container-title":["Computers, Materials & Continua"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/cdn.techscience.cn\/files\/cmc\/2025\/TSP_CMC-84-2\/TSP_CMC_63723\/TSP_CMC_63723.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,17]],"date-time":"2025-11-17T01:48:47Z","timestamp":1763344127000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.techscience.com\/cmc\/v84n2\/62876"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025]]},"references-count":39,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2025]]},"published-print":{"date-parts":[[2025]]}},"URL":"https:\/\/doi.org\/10.32604\/cmc.2025.063723","relation":{},"ISSN":["1546-2226"],"issn-type":[{"value":"1546-2226","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025]]}}}