{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,31]],"date-time":"2026-10-31T08:53:46Z","timestamp":1793436826392,"version":"4.5.0"},"reference-count":45,"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"},{"start":{"date-parts":[[2021,9,1]],"date-time":"2021-09-01T00:00:00Z","timestamp":1630454400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2021,9,1]],"date-time":"2021-09-01T00:00:00Z","timestamp":1630454400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/100000001","name":"NSF","doi-asserted-by":"publisher","award":["2007714"],"award-info":[{"award-number":["2007714"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000001","name":"NSF","doi-asserted-by":"publisher","award":["1955632"],"award-info":[{"award-number":["1955632"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]},{"name":"UC-NL","award":["LFR-18-548554"],"award-info":[{"award-number":["LFR-18-548554"]}]},{"DOI":"10.13039\/100006754","name":"Army Research Laboratory under Cooperative Agreement","doi-asserted-by":"publisher","award":["W911NF-17-2-0196"],"award-info":[{"award-number":["W911NF-17-2-0196"]}],"id":[{"id":"10.13039\/100006754","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.2021.3103920","type":"journal-article","created":{"date-parts":[[2021,8,13]],"date-time":"2021-08-13T20:11:47Z","timestamp":1628885507000},"page":"954-969","source":"Crossref","is-referenced-by-count":36,"title":["SQuARM-SGD: Communication-Efficient Momentum SGD for Decentralized Optimization"],"prefix":"10.1109","volume":"2","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-0360-2915","authenticated-orcid":false,"given":"Navjot","family":"Singh","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3544-8414","authenticated-orcid":false,"given":"Deepesh","family":"Data","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8417-5411","authenticated-orcid":false,"given":"Jemin","family":"George","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7313-9861","authenticated-orcid":false,"given":"Suhas","family":"Diggavi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/CDC.2017.8264668"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/TAC.2014.2366855"},{"key":"ref33","first-page":"1","article-title":"SlowMo: Improving communication-efficient distributed SGD with slow momentum","author":"wang","year":"2020","journal-title":"Proc Int Conf Learn Represent"},{"key":"ref32","first-page":"5381","article-title":"A unified theory of decentralized SGD with changing topology and local updates","volume":"119","author":"koloskova","year":"2020","journal-title":"Proc Int Conf Mach Learn (ICML)"},{"key":"ref31","first-page":"4151","article-title":"The marginal value of adaptive gradient methods in machine learning","author":"wilson","year":"2017","journal-title":"Proc 31st Int Conf Neural Inf Process Syst"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/ICC47138.2019.9123209"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1016\/j.automatica.2012.08.042"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/TAC.2011.2174666"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/CDC.2012.6425820"},{"key":"ref34","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":"ref10","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":"ref40","doi-asserted-by":"publisher","DOI":"10.1016\/j.automatica.2015.03.001"},{"key":"ref11","first-page":"560","article-title":"signSGD: Compressed optimisation for non-convex problems","author":"bernstein","year":"2018","journal-title":"Proc Int Conf Mach Learn (ICML)"},{"key":"ref12","first-page":"14668","article-title":"Qsparse-local-SGD: Distributed SGD with quantization, sparsification and local computations","author":"basu","year":"2019","journal-title":"Advances in neural information processing systems"},{"key":"ref13","first-page":"1","article-title":"Local SGD converges fast and communicates little","author":"stich","year":"2019","journal-title":"Proc Int Conf Learn Represent (ICLR)"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v33i01.33015693"},{"key":"ref15","article-title":"Iterative parameter mixing for distributed large-margin training of structured predictors for natural language processing","author":"coppola","year":"2015"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2018\/410"},{"key":"ref17","first-page":"5330","article-title":"Can decentralized algorithms outperform centralized algorithms? A case study for decentralized parallel stochastic gradient descent","author":"lian","year":"2017","journal-title":"Advances in neural information processing systems"},{"key":"ref18","first-page":"7663","article-title":"Communication compression for decentralized training","author":"tang","year":"2018","journal-title":"Advances in neural information processing systems"},{"key":"ref19","first-page":"3449","article-title":"SPARQ-SGD: Event-triggered and compressed communication in decentralized optimization","author":"singh","year":"2020","journal-title":"Proc 59th IEEE Conf Decis Control (CDC)"},{"key":"ref28","first-page":"7184","article-title":"On the linear speedup analysis of communication efficient momentum SGD for distributed non-convex optimization","author":"yu","year":"2019","journal-title":"Proc Int Conf Mach Learn (ICML)"},{"key":"ref4","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":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/TAC.2017.2648041"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D17-1045"},{"key":"ref6","first-page":"5973","article-title":"The convergence of sparsified gradient methods","author":"alistarh","year":"2018","journal-title":"Advances in neural information processing systems"},{"key":"ref29","author":"wang","year":"2018","journal-title":"Cooperative SGD A unified framework for the design and analysis of communication-efficient SGD algorithms"},{"key":"ref5","first-page":"4447","article-title":"Sparsified SGD with memory","author":"stich","year":"2018","journal-title":"Advances in neural information processing systems"},{"key":"ref8","first-page":"1508","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":"ref7","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":"ref2","first-page":"1488","article-title":"Scalable distributed DNN training using commodity GPU cloud computing","author":"strom","year":"2015","journal-title":"Proc Annu Conf Int Speech Commun Assoc (Interspeech)"},{"key":"ref9","first-page":"3329","article-title":"Distributed mean estimation with limited communication","author":"suresh","year":"2017","journal-title":"Proc Int Conf Mach Learn (ICML)"},{"key":"ref1","author":"kone?n?","year":"2016","journal-title":"Federated learning Strategies for improving communication efficiency"},{"key":"ref20","first-page":"1","article-title":"Decentralized deep learning with arbitrary communication compression","author":"koloskova","year":"2020","journal-title":"Proc Int Conf Learn Represent (ICLR)"},{"key":"ref45","first-page":"2074","article-title":"Learning structured sparsity in deep neural networks","author":"wen","year":"2016","journal-title":"Advances in neural information processing systems"},{"key":"ref22","first-page":"6155","article-title":"DoubleSqueeze: Parallel stochastic gradient descent with double-pass error-compensated compression","author":"tang","year":"2019","journal-title":"Proc Int Conf Mach Learn (ICML)"},{"key":"ref21","first-page":"3478","article-title":"Decentralized stochastic optimization and gossip algorithms with compressed communication","author":"koloskova","year":"2019","journal-title":"Proc Int Conf Mach Learn (ICML)"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/CDC.2018.8619311"},{"key":"ref24","author":"krizhevsky","year":"2009","journal-title":"Cifar-10"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1016\/j.automatica.2015.11.015"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref44","first-page":"3043","article-title":"Asynchronous decentralized parallel stochastic gradient descent","author":"lian","year":"2017","journal-title":"Proc Int Conf Mach Learn (ICML)"},{"key":"ref26","first-page":"344","article-title":"Stochastic gradient push for distributed deep learning","author":"assran","year":"2019","journal-title":"Proc Int Conf Mach Learn (ICML)"},{"key":"ref43","first-page":"5050","article-title":"LAG: Lazily aggregated gradient for communication-efficient distributed learning","author":"chen","year":"2018","journal-title":"Advances in neural information processing systems"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/CDC.2018.8619539"}],"container-title":["IEEE Journal on Selected Areas in Information Theory"],"original-title":[],"link":[{"URL":"https:\/\/ieeexplore.ieee.org\/ielam\/8700143\/9540917\/9513259-aam.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/8700143\/9540917\/09513259.pdf?arnumber=9513259","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,4,8]],"date-time":"2022-04-08T18:55:12Z","timestamp":1649444112000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9513259\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,9]]},"references-count":45,"journal-issue":{"issue":"3"},"URL":"https:\/\/doi.org\/10.1109\/jsait.2021.3103920","relation":{},"ISSN":["2641-8770"],"issn-type":[{"value":"2641-8770","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,9]]}}}