{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,7,2]],"date-time":"2025-07-02T04:05:39Z","timestamp":1751429139267,"version":"3.41.0"},"reference-count":56,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"7","license":[{"start":{"date-parts":[[2025,7,1]],"date-time":"2025-07-01T00:00:00Z","timestamp":1751328000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,7,1]],"date-time":"2025-07-01T00:00:00Z","timestamp":1751328000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,7,1]],"date-time":"2025-07-01T00:00:00Z","timestamp":1751328000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"Shun Hing Institute of Advanced Engineering, The Chinese University of Hong Kong","award":["#MMT-p5-23"],"award-info":[{"award-number":["#MMT-p5-23"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Automat. Contr."],"published-print":{"date-parts":[[2025,7]]},"DOI":"10.1109\/tac.2025.3545551","type":"journal-article","created":{"date-parts":[[2025,2,25]],"date-time":"2025-02-25T18:58:54Z","timestamp":1740509934000},"page":"4703-4718","source":"Crossref","is-referenced-by-count":0,"title":["Tighter Analysis for Decentralized Stochastic Gradient Method: Impact of Data Homogeneity"],"prefix":"10.1109","volume":"70","author":[{"ORCID":"https:\/\/orcid.org\/0009-0006-1024-1344","authenticated-orcid":false,"given":"Qiang","family":"Li","sequence":"first","affiliation":[{"name":"Department of Systems Engineering & Engineering Management, Faculty of Engineering, The Chinese University of Hong Kong, Hong Kong"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4796-4483","authenticated-orcid":false,"given":"Hoi-To","family":"Wai","sequence":"additional","affiliation":[{"name":"Department of Systems Engineering & Engineering Management, Faculty of Engineering, The Chinese University of Hong Kong, Hong Kong"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/TAC.2017.2705559"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/TSP.2012.2194290"},{"key":"ref3","first-page":"1626","article-title":"Finite-time analysis of distributed TD (0) with linear function approximation on multi-agent reinforcement learning","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Doan","year":"2019"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1098\/rspa.2007.1900"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/TAC.2017.2690401"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/TAC.2016.2626578"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijmedinf.2018.01.007"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2020.2970170"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2020.2975210"},{"article-title":"Can decentralized algorithms outperform centralized algorithms? A case study for decentralized parallel stochastic gradient descent","year":"2017","author":"Lian","key":"ref10"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/TAC.2021.3126253"},{"key":"ref12","first-page":"11422","article-title":"An improved analysis of gradient tracking for decentralized machine learning","volume":"34","author":"Koloskova","year":"2021","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1007\/s10107-020-01487-0"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/TSP.2021.3062553"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/TSP.2022.3184770"},{"issue":"280","key":"ref16","first-page":"1","article-title":"Removing data heterogeneity influence enhances network topology dependence of decentralized SGD","volume":"24","author":"Yuan","year":"2023","journal-title":"J. Mach. Learn. Res."},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/TAC.2022.3201141"},{"key":"ref18","first-page":"7111","article-title":"Optimal complexity in decentralized training","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Lu","year":"2021"},{"key":"ref19","first-page":"36382","article-title":"Revisiting optimal convergence rate for smooth and non-convex stochastic decentralized optimization","volume":"35","author":"Yuan","year":"2022","journal-title":"Adv. Neural Informat. Proces. Syst."},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/cdc.1984.272358"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1007\/s10957-010-9737-7"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/TAC.2012.2209984"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1137\/16M1084316"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/TSIPN.2016.2524588"},{"key":"ref25","first-page":"4848","article-title":"$ d^{2}$: Decentralized training over decentralized data","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Tang","year":"2018"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/TSP.2018.2875898"},{"key":"ref27","first-page":"344","article-title":"Stochastic gradient push for distributed deep learning","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Assran","year":"2019"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2016.2593769"},{"key":"ref29","article-title":"Can decentralized algorithms outperform centralized algorithms? A case study for decentralized parallel stochastic gradient descent","volume":"30","author":"Lian","year":"125","journal-title":"Adv. Neural Inform. Process. Syst."},{"issue":"355","key":"ref30","first-page":"1","article-title":"Beyond spectral gap: The role of the topology in decentralized learning","volume":"24","author":"Vogels","year":"2023","journal-title":"J. Mach. Learn. Res."},{"key":"ref31","first-page":"1","article-title":"Moshpit SGD: Communication-efficient decentralized training on heterogeneous unreliable devices","volume":"34","author":"Ryabinin","year":"2021","journal-title":"Adv. Neural Inform. Proces. Syst."},{"key":"ref32","first-page":"4203","article-title":"Statistically preconditioned accelerated gradient method for distributed optimization","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Hendrikx","year":"2020"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/CDC51059.2022.9992403"},{"key":"ref34","first-page":"8172","article-title":"Distributed saddle-point problems under data similarity","volume":"34","author":"Beznosikov","year":"2021","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1137\/19M1259973"},{"key":"ref36","first-page":"5721","article-title":"Acceleration in distributed optimization under similarity","volume-title":"Proc. Int. Conf. Artif. Intell. Statist.","author":"Tian","year":"2022"},{"volume-title":"Reversible Markov Chains and Random Walks on Graphs","year":"1995","author":"Aldous","key":"ref37"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1137\/120880811"},{"key":"ref39","first-page":"5200","article-title":"SGD: General analysis and improved rates","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Gower","year":"2019"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1007\/s10107-022-01822-7"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2011.2182178"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2020.2975212"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/TAC.2008.2009515"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1007\/s10589-020-00183-1"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1137\/16M1080173"},{"key":"ref46","volume-title":"Introductory Lectures on Convex Optimization: A Basic Course","volume":"87","author":"Nesterov","year":"2003"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1007\/BF00115009"},{"key":"ref48","first-page":"5039","article-title":"Analysis of stochastic processes through replay buffers","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Di-Castro","year":"2022"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1109\/LCSYS.2023.3287952"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1287\/opre.2020.2024"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v32i1.12079"},{"key":"ref52","first-page":"2803","article-title":"Finite-time error bounds for linear stochastic approximation and TD learning","volume-title":"Proc. 32nd Ann. Conf. Learn. Theory","author":"Srikant","year":"2019"},{"key":"ref53","first-page":"2260","article-title":"Momentum improves normalized SGD","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Cutkosky","year":"2020"},{"key":"ref54","first-page":"1192","article-title":"Sharp analysis for nonconvex SGD escaping from saddle points","volume-title":"Proc. 32nd Ann. Conf. Learn. Theory","author":"Fang","year":"2019"},{"key":"ref55","first-page":"1672","article-title":"Refined convergence and topology learning for decentralized SGD with heterogeneous data","volume-title":"Proc. Int. Conf. Artif. Intell. Statist.","author":"Le Bars","year":"2023"},{"key":"ref56","first-page":"38449","article-title":"Multi-agent performative prediction with greedy deployment and consensus seeking agents","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Li","year":"2022"}],"container-title":["IEEE Transactions on Automatic Control"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/9\/11060003\/10904000.pdf?arnumber=10904000","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,1]],"date-time":"2025-07-01T05:33:11Z","timestamp":1751347991000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10904000\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7]]},"references-count":56,"journal-issue":{"issue":"7"},"URL":"https:\/\/doi.org\/10.1109\/tac.2025.3545551","relation":{},"ISSN":["0018-9286","1558-2523","2334-3303"],"issn-type":[{"type":"print","value":"0018-9286"},{"type":"electronic","value":"1558-2523"},{"type":"electronic","value":"2334-3303"}],"subject":[],"published":{"date-parts":[[2025,7]]}}}