{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,10]],"date-time":"2026-07-10T16:34:21Z","timestamp":1783701261482,"version":"3.55.0"},"reference-count":60,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"12","license":[{"start":{"date-parts":[[2022,12,1]],"date-time":"2022-12-01T00:00:00Z","timestamp":1669852800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2022,12,1]],"date-time":"2022-12-01T00:00:00Z","timestamp":1669852800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2022,12,1]],"date-time":"2022-12-01T00:00:00Z","timestamp":1669852800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100003130","name":"Research Foundation\u2013Flanders (FWO) Research","doi-asserted-by":"publisher","award":["G093817N"],"award-info":[{"award-number":["G093817N"]}],"id":[{"id":"10.13039\/501100003130","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Neural Netw. Learning Syst."],"published-print":{"date-parts":[[2022,12]]},"DOI":"10.1109\/tnnls.2021.3084806","type":"journal-article","created":{"date-parts":[[2021,6,10]],"date-time":"2021-06-10T19:44:52Z","timestamp":1623354292000},"page":"7330-7344","source":"Crossref","is-referenced-by-count":63,"title":["Learned Gradient Compression for Distributed Deep Learning"],"prefix":"10.1109","volume":"33","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-3232-3029","authenticated-orcid":false,"given":"Lusine","family":"Abrahamyan","sequence":"first","affiliation":[{"name":"Department of Electronics and Informatics (ETRO), Vrije Universiteit Brussel, Brussels, Belgium"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yiming","family":"Chen","sequence":"additional","affiliation":[{"name":"Department of Electronics and Informatics (ETRO), Vrije Universiteit Brussel, Brussels, Belgium"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3377-2675","authenticated-orcid":false,"given":"Giannis","family":"Bekoulis","sequence":"additional","affiliation":[{"name":"Department of Electronics and Informatics (ETRO), Vrije Universiteit Brussel, Brussels, Belgium"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9300-5860","authenticated-orcid":false,"given":"Nikos","family":"Deligiannis","sequence":"additional","affiliation":[{"name":"Department of Electronics and Informatics (ETRO), Vrije Universiteit Brussel, Brussels, Belgium"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.1976.1055508"},{"key":"ref38","article-title":"Bidirectional compression in heterogeneous settings for distributed or federated learning with partial participation: Tight convergence guarantees","author":"philippenko","year":"2020","journal-title":"arXiv 2006 14591"},{"key":"ref33","article-title":"Federated learning with quantized global model updates","author":"amiri","year":"2020","journal-title":"arXiv 2006 10672"},{"key":"ref32","article-title":"Very deep convolutional networks for large-scale image recognition","author":"simonyan","year":"2015","journal-title":"Proc 3rd Int Conf Learn Represent (ICLR)"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2019.2944481"},{"key":"ref30","year":"2017","journal-title":"Federated learning Collaborative machine learning without centralized training data"},{"key":"ref37","first-page":"8253","article-title":"FetchSGD: Communication-efficient federated learning with sketching","author":"rothchild","year":"2020","journal-title":"Proc 37th Int Conf Mach Learn"},{"key":"ref36","first-page":"4447","article-title":"Sparsified sgd with memory","author":"stich","year":"2018","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref35","first-page":"11446","article-title":"Communication-efficient distributed blockwise momentum SGD with error-feedback","author":"zheng","year":"2019","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref34","first-page":"3252","article-title":"Error feedback fixes SignSGD and other gradient compression schemes","author":"karimireddy","year":"2019","journal-title":"Proc Int Conf Mach Learn (ICML)"},{"key":"ref60","first-page":"278","author":"furht","year":"2006","journal-title":"Huffman Coding"},{"key":"ref28","first-page":"1097","article-title":"ImageNet classification with deep convolutional neural networks","author":"krizhevsky","year":"2012","journal-title":"Proc Adv Neural Inf Process Syst (NIPS)"},{"key":"ref27","first-page":"1299","article-title":"Gradient sparsification for communication-efficient distributed optimization","author":"wangni","year":"2018","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref29","article-title":"Learning multiple layers of features from tiny images","author":"krizhevsky","year":"2009"},{"key":"ref2","first-page":"1","article-title":"BERT: Pre-training of deep bidirectional transformers for language understanding","author":"devlin","year":"2019","journal-title":"Proc Conf North Amer Chapter Assoc Comput Linguistics"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01216-8_12"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/PROC.1967.5493"},{"key":"ref22","first-page":"1","article-title":"Communication efficient distributed machine learning with the parameter server","author":"li","year":"2014","journal-title":"Proc Adv Neural Inf Process Syst (NIPS)"},{"key":"ref21","first-page":"13551","article-title":"ScaleCom: Scalable sparsified gradient compression for communication-efficient distributed training","author":"chen","year":"2020","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-540-24685-5_1"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1016\/j.jpdc.2008.09.002"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/5.726791"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D17-1045"},{"key":"ref50","article-title":"Automatic differentiation in PyTorch","author":"paszke","year":"2017","journal-title":"Proc NIPS Workshop Autodiff"},{"key":"ref51","first-page":"1509","article-title":"TernGrad: Ternary gradients to reduce communication in distributed deep learning","author":"wen","year":"2017","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1145\/3426745.3431334"},{"key":"ref58","first-page":"7","article-title":"Tiny ImageNet visual recognition challenge","volume":"7","author":"le","year":"2015","journal-title":"CS 231N"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-10599-4_29"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v31i1.11231"},{"key":"ref55","article-title":"Accurate, large minibatch SGD: Training ImageNet in 1 hour","author":"goyal","year":"2017","journal-title":"arXiv 1706 02677"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-24449-0_8"},{"key":"ref53","first-page":"14695","article-title":"Qsparse-local-SGD: Distributed SGD with quantization, sparsification and local computations","author":"basu","year":"2019","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref52","first-page":"5973","article-title":"The convergence of sparsified gradient methods","author":"alistarh","year":"2018","journal-title":"Proc NIPS"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01234-2_48"},{"key":"ref11","first-page":"9","article-title":"Distilling the knowledge in a neural network","volume":"1050","author":"hinton","year":"2015","journal-title":"Stat"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/18.490552"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1145\/1150402.1150464"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2015-354"},{"key":"ref14","first-page":"1","article-title":"Deep gradient compression: Reducing the communication bandwidth for distributed training","author":"lin","year":"2018","journal-title":"Proc Int Conf Learn Represent (ICLR)"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2014-274"},{"key":"ref16","first-page":"1707","article-title":"QSGD: Communication-efficient SGD via gradient quantization and encoding","author":"alistarh","year":"2017","journal-title":"Adv Neural Inf Process Syst (NIPS)"},{"key":"ref17","article-title":"Nested dithered quantization for communication reduction in distributed training","author":"abdi","year":"2019","journal-title":"arXiv 1904 01197"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/SPAWC.2019.8815453"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.1973.1055037"},{"key":"ref4","article-title":"Qsparse-local-SGD: Distributed SGD with quantization, sparsification, and local computations","author":"basu","year":"2019","journal-title":"arXiv 1906 02367"},{"key":"ref3","first-page":"1","article-title":"Large scale distributed deep networks","author":"dean","year":"2012","journal-title":"Proc Adv Neural Inf Process Syst (NIPS)"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref8","article-title":"Quantizing deep convolutional networks for efficient inference: A whitepaper","author":"krishnamoorthi","year":"2018","journal-title":"arXiv 1806 08342"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00286"},{"key":"ref49","author":"katz","year":"1991","journal-title":"String Searcher and Compressor Using Same"},{"key":"ref9","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":"ref46","doi-asserted-by":"publisher","DOI":"10.1016\/j.patrec.2008.04.005"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.660"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1989.1.4.541"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.123"},{"key":"ref42","article-title":"DRASIC: Distributed recurrent autoencoder for scalable image compression","author":"diao","year":"2019","journal-title":"arXiv 1903 09887"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2002.808103"},{"key":"ref44","first-page":"6850","article-title":"Deep compressed sensing","author":"wu","year":"2019","journal-title":"Proc Int Conf Mach Learn (ICML)"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/DCC.2018.00023"}],"container-title":["IEEE Transactions on Neural Networks and Learning Systems"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/5962385\/9966944\/09451554.pdf?arnumber=9451554","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,12,19]],"date-time":"2022-12-19T19:53:56Z","timestamp":1671479636000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9451554\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,12]]},"references-count":60,"journal-issue":{"issue":"12"},"URL":"https:\/\/doi.org\/10.1109\/tnnls.2021.3084806","relation":{},"ISSN":["2162-237X","2162-2388"],"issn-type":[{"value":"2162-237X","type":"print"},{"value":"2162-2388","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,12]]}}}