{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,9,3]],"date-time":"2025-09-03T10:22:39Z","timestamp":1756894959730,"version":"3.37.3"},"reference-count":45,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"2","license":[{"start":{"date-parts":[[2024,2,1]],"date-time":"2024-02-01T00:00:00Z","timestamp":1706745600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2024,2,1]],"date-time":"2024-02-01T00:00:00Z","timestamp":1706745600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2024,2,1]],"date-time":"2024-02-01T00:00:00Z","timestamp":1706745600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"National Science and Engineering Research Council of Canada through a Discovery Grant"},{"name":"MOST","award":["111-2221-E-A49-068-MY3"],"award-info":[{"award-number":["111-2221-E-A49-068-MY3"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Commun."],"published-print":{"date-parts":[[2024,2]]},"DOI":"10.1109\/tcomm.2023.3327778","type":"journal-article","created":{"date-parts":[[2023,10,26]],"date-time":"2023-10-26T17:54:13Z","timestamp":1698342853000},"page":"845-860","source":"Crossref","is-referenced-by-count":3,"title":["M22: A Communication-Efficient Algorithm for Federated Learning Inspired by Rate-Distortion"],"prefix":"10.1109","volume":"72","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-7366-724X","authenticated-orcid":false,"given":"Yangyi","family":"Liu","sequence":"first","affiliation":[{"name":"Department of Electrical and Computer Engineering, McMaster University, Hamilton, ON, Canada"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Stefano","family":"Rini","sequence":"additional","affiliation":[{"name":"Department of Electrical and Computer Engineering, National Yang Ming Chiao Tung University (NYCU), Hsinchu, Taiwan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Sadaf","family":"Salehkalaibar","sequence":"additional","affiliation":[{"name":"Department of Electrical and Computer Engineering, University of Toronto, Toronto, ON, Canada"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8084-9332","authenticated-orcid":false,"given":"Jun","family":"Chen","sequence":"additional","affiliation":[{"name":"Department of Electrical and Computer Engineering, McMaster University, Hamilton, ON, Canada"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Federated learning: Strategies for improving communication efficiency","author":"Kone\u010dn\u00fd","year":"2016","journal-title":"arXiv:1610.05492"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2020.2975749"},{"volume-title":"Parallel and Distributed Computation: Numerical Methods","year":"2015","author":"Bertsekas","key":"ref3"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1137\/090759574"},{"key":"ref5","first-page":"9850","article-title":"ATOMO: Communication-efficient learning via atomic sparsification","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Wang"},{"key":"ref6","first-page":"5973","article-title":"The convergence of sparsified gradient methods","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Alistarh"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/ISIT44484.2020.9174245"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2019.2944481"},{"key":"ref9","article-title":"Expanding the reach of federated learning by reducing client resource requirements","author":"Caldas","year":"2018","journal-title":"arXiv:1812.07210"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2021.3131614"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/JSAC.2021.3118428"},{"key":"ref12","first-page":"4448","article-title":"Sparsified SGD with memory","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Stich"},{"key":"ref13","article-title":"Variance-based gradient compression for efficient distributed deep learning","author":"Tsuzuku","year":"2018","journal-title":"arXiv:1802.06058"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/d17-1045"},{"article-title":"Deep gradient compression: Reducing the communication bandwidth for distributed training","volume-title":"Proc. Int. Conf. Learn. Represent. (ICLR)","author":"Lin","key":"ref15"},{"key":"ref16","first-page":"3973","article-title":"FedBoost: A communication-efficient algorithm for federated learning","volume-title":"Proc. 37th Int. Conf. Mach. Learn.","volume":"119","author":"Hamer"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2014-274"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/tit.2022.3161620"},{"key":"ref19","first-page":"560","article-title":"signSGD: Compressed optimization for non-convex problems","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Bernstein"},{"key":"ref20","first-page":"7564","article-title":"cpSGD: Communication-efficient and differentially-private distributed SGD","volume-title":"Proc. Adv. Neural Inf. Process. Syst. (NIPS)","author":"Agarwal"},{"key":"ref21","article-title":"Privacy for free: Communication-efficient learning with differential privacy using sketches","author":"Li","year":"2019","journal-title":"arXiv:1911.00972"},{"key":"ref22","first-page":"13142","article-title":"Communication-efficient distributed SGD with sketching","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"32","author":"Ivkin"},{"key":"ref23","first-page":"8253","article-title":"FetchSGD: Communication-efficient federated learning with sketching","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Rothchild"},{"key":"ref24","first-page":"1","article-title":"QSGD: Communication-efficient sgd via gradient quantization and encoding","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"30","author":"Alistarh"},{"key":"ref25","first-page":"1509","article-title":"TernGrad: Ternary gradients to reduce communication in distributed deep learning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Wen"},{"article-title":"Learning differentially private recurrent language models","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"McMahan","key":"ref26"},{"key":"ref27","article-title":"Concentrated differentially private and utility preserving federated learning","author":"Hu","year":"2020","journal-title":"arXiv:2003.13761"},{"key":"ref28","first-page":"4900","article-title":"Hybrid 8-bit floating point (HFP8) training and inference for deep neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"32","author":"Sun"},{"key":"ref29","first-page":"3088","article-title":"Approximate inference turns deep networks into Gaussian processes","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Khan"},{"key":"ref30","article-title":"An information-theoretic justification for model pruning","author":"Isik","year":"2021","journal-title":"arXiv:2102.08329"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/ICC45855.2022.9838754"},{"key":"ref32","first-page":"3304","article-title":"Don\u2019t waste your bits! Squeeze activations and gradients for deep neural networks via TinyScript","volume-title":"Proc. 37th Int. Conf. Mach. Learn. (ICML)","volume":"119","author":"Fu"},{"key":"ref33","first-page":"1273","article-title":"Communication-efficient learning of deep networks from decentralized data","volume-title":"Artificial Intelligence and Statistics","volume":"54","author":"McMahan","year":"2017"},{"volume-title":"The Data Compression","year":"1995","author":"Nelson","key":"ref34"},{"key":"ref35","article-title":"A mean-field limit for certain deep neural networks","author":"Ara\u00fajo","year":"2019","journal-title":"arXiv:1906.00193"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/TCOM.1980.1094577"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.1806579115"},{"key":"ref38","article-title":"A rigorous framework for the mean field limit of multilayer neural networks","author":"Nguyen","year":"2020","journal-title":"arXiv:2001.11443"},{"key":"ref39","article-title":"Deep neural networks as Gaussian processes","author":"Lee","year":"2017","journal-title":"arXiv:1711.00165"},{"key":"ref40","article-title":"Gaussian process behaviour in wide deep neural networks","author":"de G. Matthews","year":"2018","journal-title":"arXiv:1804.11271"},{"key":"ref41","first-page":"7662","article-title":"Communication compression for decentralized training","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Tang"},{"key":"ref42","first-page":"1","article-title":"Layer-adaptive sparsity for the magnitude-based pruning","volume-title":"Proc. ICLR","author":"Lee"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.1968.1054193"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1109\/JRPROC.1951.230419"},{"key":"ref45","article-title":"Fashion-MNIST: A novel image dataset for benchmarking machine learning algorithms","author":"Xiao","year":"2017","journal-title":"arXiv:1708.07747"}],"container-title":["IEEE Transactions on Communications"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/26\/10438109\/10296880.pdf?arnumber=10296880","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,3,2]],"date-time":"2024-03-02T11:06:10Z","timestamp":1709377570000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10296880\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,2]]},"references-count":45,"journal-issue":{"issue":"2"},"URL":"https:\/\/doi.org\/10.1109\/tcomm.2023.3327778","relation":{},"ISSN":["0090-6778","1558-0857"],"issn-type":[{"type":"print","value":"0090-6778"},{"type":"electronic","value":"1558-0857"}],"subject":[],"published":{"date-parts":[[2024,2]]}}}