{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,20]],"date-time":"2026-01-20T02:52:12Z","timestamp":1768877532537,"version":"3.49.0"},"reference-count":28,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"3","license":[{"start":{"date-parts":[[2021,9,1]],"date-time":"2021-09-01T00:00:00Z","timestamp":1630454400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2021,9,1]],"date-time":"2021-09-01T00:00:00Z","timestamp":1630454400000},"content-version":"am","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"}],"funder":[{"DOI":"10.13039\/501100003816","name":"Huawei Technologies Canada","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100003816","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100000038","name":"Natural Science and Engineering Research Council (NSERC) of Canada through a Discovery Research Grant","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100000038","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE J. Sel. Areas Inf. Theory"],"published-print":{"date-parts":[[2021,9]]},"DOI":"10.1109\/jsait.2021.3103494","type":"journal-article","created":{"date-parts":[[2021,8,11]],"date-time":"2021-08-11T20:14:46Z","timestamp":1628712886000},"page":"970-986","source":"Crossref","is-referenced-by-count":3,"title":["Compressing Gradients by Exploiting Temporal Correlation in Momentum-SGD"],"prefix":"10.1109","volume":"2","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-7628-7568","authenticated-orcid":false,"given":"Tharindu B.","family":"Adikari","sequence":"first","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8100-5599","authenticated-orcid":false,"given":"Stark C.","family":"Draper","sequence":"additional","affiliation":[]}],"member":"263","reference":[{"key":"ref10","first-page":"1545","article-title":"Predictive coding for lossless dataset compression","author":"barowsky","year":"2021","journal-title":"Proc Int Conf Acoust Speech Signal Process"},{"key":"ref11","author":"wiegand","year":"2011","journal-title":"Source Coding Part I of Fundamentals of Source and Video Coding"},{"key":"ref12","first-page":"1488","article-title":"Scalable distributed DNN training using commodity GPU cloud computing","author":"strom","year":"2015","journal-title":"Proc Conf Int Speech Commun Assoc"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/MLHPC.2016.004"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D17-1045"},{"key":"ref15","author":"lin","year":"2017","journal-title":"Deep Gradient Compression Reducing the Communication Bandwidth for Distributed Training"},{"key":"ref16","first-page":"1299","article-title":"Gradient sparsification for communication-efficient distributed optimization","author":"wangni","year":"2018","journal-title":"Advances in neural information processing systems"},{"key":"ref17","first-page":"4447","article-title":"Sparsified SGD with memory","author":"stich","year":"2018","journal-title":"Advances in neural information processing systems"},{"key":"ref18","article-title":"ATOMO: Communication-efficient learning via atomic sparsification","author":"wang","year":"2018","journal-title":"Advances in Neural IInformation Processing Systems"},{"key":"ref19","first-page":"4035","article-title":"ZipML: Training linear models with end-to-end low precision, and a little bit of deep learning","author":"zhang","year":"2017","journal-title":"Proc Int Conf Mach Learn"},{"key":"ref28","author":"sergeev","year":"2018","journal-title":"Horovod fast and easy distributed deep learning in tensorflow"},{"key":"ref4","author":"devlin","year":"2018","journal-title":"BERT Pre-training of deep bidirectional transformers for language understanding"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/IJCNN.2019.8852172"},{"key":"ref3","article-title":"Deep gradient compression: Reducing the communication bandwidth for distributed training","author":"lin","year":"2018","journal-title":"Proc Int Conf Learn Represent"},{"key":"ref6","first-page":"559","article-title":"SignSGD: Compressed optimisation for non-convex problems","author":"bernstein","year":"2018","journal-title":"Proc Int Conf Mach Learn"},{"key":"ref5","first-page":"1509","article-title":"TernGrad: Ternary gradients to reduce communication in distributed deep learning","author":"wen","year":"2017","journal-title":"Advances in neural information processing systems"},{"key":"ref8","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"},{"key":"ref7","first-page":"5321","article-title":"Error compensated quantized SGD and its applications to large-scale distributed optimization","author":"wu","year":"2018","journal-title":"Proc Int Conf Mach Learn"},{"key":"ref2","first-page":"1058","article-title":"1-bit stochastic gradient descent and application to data-parallel distributed training of speech DNNs","author":"seide","year":"2014","journal-title":"Proc Conf Int Speech Commun Assoc"},{"key":"ref9","first-page":"11446","article-title":"Communication-efficient distributed blockwise momentum SGD with error-feedback","author":"zheng","year":"2019","journal-title":"Advances in neural information processing systems"},{"key":"ref1","first-page":"1709","article-title":"QSGD: Communication-efficient SGD via gradient quantization and encoding","author":"alistarh","year":"2017","journal-title":"Advances in neural information processing systems"},{"key":"ref20","first-page":"249","article-title":"Understanding the difficulty of training deep feedforward neural networks","author":"glorot","year":"2010","journal-title":"Proc Int Conf Artif Intell Stat"},{"key":"ref22","author":"yu","year":"2018","journal-title":"GradiVeQ Vector quantization for bandwidth-efficient gradient aggregation in distributed CNN training"},{"key":"ref21","author":"shi","year":"2019","journal-title":"Understanding Top-k sparsification in distributed deep learning"},{"key":"ref24","author":"zagoruyko","year":"2016","journal-title":"Wide residual networks"},{"key":"ref23","author":"oppenheim","year":"2001","journal-title":"Discrete-Time Signal Processing"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/TASSP.1986.1164905"},{"key":"ref25","author":"chrabaszcz","year":"2017","journal-title":"A downsampled variant of imagenet as an alternative to the cifar datasets"}],"container-title":["IEEE Journal on Selected Areas in Information Theory"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/8700143\/9540917\/09511618.pdf?arnumber=9511618","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,11,17]],"date-time":"2021-11-17T23:17:16Z","timestamp":1637191036000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9511618\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,9]]},"references-count":28,"journal-issue":{"issue":"3"},"URL":"https:\/\/doi.org\/10.1109\/jsait.2021.3103494","relation":{},"ISSN":["2641-8770"],"issn-type":[{"value":"2641-8770","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,9]]}}}