{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,7]],"date-time":"2026-03-07T18:04:36Z","timestamp":1772906676955,"version":"3.50.1"},"reference-count":58,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","license":[{"start":{"date-parts":[[2021,1,1]],"date-time":"2021-01-01T00:00:00Z","timestamp":1609459200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"}],"funder":[{"DOI":"10.13039\/501100010418","name":"Institute for Information and Communications Technology Promotion through the Korean Government","doi-asserted-by":"publisher","award":["1711073912"],"award-info":[{"award-number":["1711073912"]}],"id":[{"id":"10.13039\/501100010418","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100002628","name":"Incheon National University, South Korea","doi-asserted-by":"publisher","award":["2019"],"award-info":[{"award-number":["2019"]}],"id":[{"id":"10.13039\/501100002628","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Access"],"published-print":{"date-parts":[[2021]]},"DOI":"10.1109\/access.2021.3067889","type":"journal-article","created":{"date-parts":[[2021,3,22]],"date-time":"2021-03-22T20:22:09Z","timestamp":1616444529000},"page":"47194-47203","source":"Crossref","is-referenced-by-count":18,"title":["Training Multi-Bit Quantized and Binarized Networks with a Learnable Symmetric Quantizer"],"prefix":"10.1109","volume":"9","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6388-1883","authenticated-orcid":false,"given":"Phuoc","family":"Pham","sequence":"first","affiliation":[{"name":"Department of Electronics Engineering, Incheon National University, Incheon, Republic of Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5336-5631","authenticated-orcid":false,"given":"Jacob A.","family":"Abraham","sequence":"additional","affiliation":[{"name":"Computer Engineering Research Center, The University of Texas at Austin, Austin, TX, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5819-1995","authenticated-orcid":false,"given":"Jaeyong","family":"Chung","sequence":"additional","affiliation":[{"name":"Department of Electronics Engineering, Incheon National University, Incheon, Republic of Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref39","article-title":"Apprentice: Using knowledge distillation techniques to improve low-precision network accuracy","author":"mishra","year":"2017","journal-title":"arXiv 1711 05852"},{"key":"ref38","article-title":"SGDR: Stochastic gradient descent with warm restarts","author":"loshchilov","year":"2017","journal-title":"Proc 5th Int Conf Learn Represent (ICLR)"},{"key":"ref33","first-page":"2849","article-title":"Fixed point quantization of deep convolutional networks","author":"lin","year":"2016","journal-title":"Proc Int Conf Mach Learn"},{"key":"ref32","article-title":"Additive powers-of-two quantization: An efficient non-uniform discretization for neural networks","author":"li","year":"2019","journal-title":"arXiv 1909 13144"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1145\/3065386"},{"key":"ref30","article-title":"Learning multiple layers of features from tiny images","author":"krizhevsky","year":"2009"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01267-0_44"},{"key":"ref36","article-title":"AQD: Towards accurate quantized object detection","author":"chen","year":"2020","journal-title":"arXiv 2007 06919"},{"key":"ref35","first-page":"345","article-title":"Towards accurate binary convolutional neural network","author":"lin","year":"2017","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref34","article-title":"Rotated binary neural network","author":"lin","year":"2020","journal-title":"Proc Annu Conf Neural Inf Process Syst (NeurIPS)"},{"key":"ref28","article-title":"BinaryDuo: Reducing gradient mismatch in binary activation network by coupling binary activations","author":"kim","year":"2020","journal-title":"Proc 8th Int Conf Learn Represent (ICLR)"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58610-2_34"},{"key":"ref29","article-title":"QKD: Quantization-aware knowledge distillation","author":"kim","year":"2019","journal-title":"arXiv 1911 12491"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW50498.2020.00356"},{"key":"ref1","article-title":"An empirical study of binary neural networks’ optimisation","author":"alizadeh","year":"2019","journal-title":"Proc 7th Int Conf Learn Represent (ICLR)"},{"key":"ref20","first-page":"223","article-title":"ProxyBNN: Learning binarized neural networks via proxy matrices","volume":"2","author":"he","year":"2020","journal-title":"Proc Eur Conf Comput Vis"},{"key":"ref22","article-title":"MobileNets: Efficient convolutional neural networks for mobile vision applications","author":"howard","year":"2017","journal-title":"arXiv 1704 04861"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00140"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00286"},{"key":"ref23","first-page":"4107","article-title":"Binarized neural networks","author":"hubara","year":"2016","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref26","article-title":"Neural network quantization with scale-adjusted training","author":"jin","year":"2020","journal-title":"Proc 31st Brit Mach Vis Conf (BMVC)"},{"key":"ref25","article-title":"Trained quantization thresholds for accurate and efficient fixed-point inference of deep neural networks","author":"jain","year":"2019","journal-title":"arXiv 1903 08066"},{"key":"ref50","article-title":"Mixed precision DNNs: All you need is a good parametrization","volume":"2","author":"uhlich","year":"2019","journal-title":"arXiv 1905 11452"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1145\/3020078.3021744"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00050"},{"key":"ref57","article-title":"DoReFa-net: Training low bitwidth convolutional neural networks with low bitwidth gradients","author":"zhou","year":"2016","journal-title":"arXiv 1606 06160 [cs]"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01237-3_23"},{"key":"ref55","article-title":"Distillation guided residual learning for binary convolutional neural networks","author":"ye","year":"2020","journal-title":"arXiv 2007 05223"},{"key":"ref54","article-title":"Binarized neural networks on the ImageNet classification task","author":"wu","year":"2016","journal-title":"arXiv 1604 03058"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00881"},{"key":"ref52","first-page":"5998","article-title":"Attention is all you need","author":"vaswani","year":"2017","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.01167"},{"key":"ref11","article-title":"Learned step size quantization","author":"esser","year":"2020","journal-title":"Proc 8th Int Conf Learn Represent (ICLR)"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58539-6_26"},{"key":"ref12","article-title":"Learned step size quantization","author":"esser","year":"2019","journal-title":"arXiv 1902 08153"},{"key":"ref13","article-title":"Gemmini: An agile systolic array generator enabling systematic evaluations of deep-learning architectures","author":"genc","year":"2019","journal-title":"arXiv 1911 09925"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00448"},{"key":"ref15","first-page":"249","article-title":"Understanding the difficulty of training deep feedforward neural networks","author":"glorot","year":"2010","journal-title":"Proc 13th Int Conf Artif Intell Statist"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00495"},{"key":"ref17","first-page":"1135","article-title":"Learning both weights and connections for efficient neural network","author":"han","year":"2015","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.123"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref4","article-title":"Accurate and efficient 2-bit quantized neural networks","author":"choi","year":"2019","journal-title":"Proc Mach Learn Syst (MLSys)"},{"key":"ref3","article-title":"XNOR-net++: Improved binary neural networks","author":"bulat","year":"2019","journal-title":"arXiv 1909 13863"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2018.2857019"},{"key":"ref5","article-title":"Pact: Parameterized clipping activation for quantized neural networks","author":"choi","year":"2018","journal-title":"arXiv 1805 06085"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/TASL.2011.2134090"},{"key":"ref7","first-page":"3123","article-title":"Binaryconnect: Training deep neural networks with binary weights during propagations","author":"courbariaux","year":"2015","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref49","first-page":"6105","article-title":"EfficientNet: Rethinking model scaling for convolutional neural networks","volume":"97","author":"tan","year":"2019","journal-title":"Mach Learn Res"},{"key":"ref9","article-title":"BNN+: Regularized binary network training","author":"darabi","year":"2018","journal-title":"arXiv 1812 11800"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00474"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-015-0816-y"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00293"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/JPROC.2017.2761740"},{"key":"ref42","article-title":"Model compression via distillation and quantization","author":"polino","year":"2018","journal-title":"arXiv 1802 05668"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01343"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46493-0_32"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00232"}],"container-title":["IEEE Access"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/6287639\/9312710\/09383003.pdf?arnumber=9383003","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,9,8]],"date-time":"2022-09-08T20:17:05Z","timestamp":1662668225000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9383003\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021]]},"references-count":58,"URL":"https:\/\/doi.org\/10.1109\/access.2021.3067889","relation":{},"ISSN":["2169-3536"],"issn-type":[{"value":"2169-3536","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021]]}}}