{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,1,21]],"date-time":"2025-01-21T05:22:02Z","timestamp":1737436922416,"version":"3.33.0"},"reference-count":66,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","license":[{"start":{"date-parts":[[2025,1,1]],"date-time":"2025-01-01T00:00:00Z","timestamp":1735689600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"funder":[{"name":"Operational Program Competitiveness, Entrepreneurship and Innovation, under the call RESEARCH ? CREATE ? INNOVATE","award":["T2EDK-01231"],"award-info":[{"award-number":["T2EDK-01231"]}]},{"DOI":"10.13039\/501100008530","name":"European Regional Development Fund","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100008530","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Access"],"published-print":{"date-parts":[[2025]]},"DOI":"10.1109\/access.2025.3528248","type":"journal-article","created":{"date-parts":[[2025,1,10]],"date-time":"2025-01-10T20:47:04Z","timestamp":1736542024000},"page":"9510-9523","source":"Crossref","is-referenced-by-count":0,"title":["Strategy-Switch: From All-Reduce to Parameter Server for Faster Efficient Training"],"prefix":"10.1109","volume":"13","author":[{"ORCID":"https:\/\/orcid.org\/0009-0000-0931-1479","authenticated-orcid":false,"given":"Nikodimos","family":"Provatas","sequence":"first","affiliation":[{"name":"School of Electrical and Computer Engineering, National Technical University of Athens, Athens, Greece"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Iasonas","family":"Chalas","sequence":"additional","affiliation":[{"name":"School of Electrical and Computer Engineering, National Technical University of Athens, Athens, Greece"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7142-8106","authenticated-orcid":false,"given":"Ioannis","family":"Konstantinou","sequence":"additional","affiliation":[{"name":"Department of Informatics and Telecommunications, University of Thessaly, Lamia, Greece"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Nectarios","family":"Koziris","sequence":"additional","affiliation":[{"name":"School of Electrical and Computer Engineering, National Technical University of Athens, Athens, Greece"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","first-page":"1","article-title":"Exploiting data distribution in distributed learning of deep classification models under the parameter server architecture","volume-title":"Proc. VLDB PhD Workshop (VLDB-PhD)","author":"Provatas"},{"key":"ref2","first-page":"1","article-title":"Convolutional-recursive deep learning for 3D object classification","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"25","author":"Socher"},{"doi-asserted-by":"publisher","key":"ref3","DOI":"10.1109\/CVPR.2017.683"},{"doi-asserted-by":"publisher","key":"ref4","DOI":"10.1109\/TGRS.2018.2815613"},{"doi-asserted-by":"publisher","key":"ref5","DOI":"10.1109\/ICMLA.2018.00179"},{"doi-asserted-by":"publisher","key":"ref6","DOI":"10.1145\/3329784"},{"doi-asserted-by":"publisher","key":"ref7","DOI":"10.1109\/TPAMI.2021.3059968"},{"doi-asserted-by":"publisher","key":"ref8","DOI":"10.1109\/ICASSP.2013.6639140"},{"doi-asserted-by":"publisher","key":"ref9","DOI":"10.1145\/3178115"},{"doi-asserted-by":"publisher","key":"ref10","DOI":"10.1109\/ICASSP.2017.7952625"},{"doi-asserted-by":"publisher","key":"ref11","DOI":"10.1109\/ICMLA.2017.0-134"},{"doi-asserted-by":"publisher","key":"ref12","DOI":"10.1145\/3077136.3080834"},{"doi-asserted-by":"publisher","key":"ref13","DOI":"10.1145\/3209280.3209526"},{"doi-asserted-by":"publisher","key":"ref14","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref15","first-page":"1","article-title":"Imagenet classification with deep convolutional neural networks","volume-title":"Proc. Adv. neural Inf. Process. Syst.","volume":"25","author":"Krizhevsky"},{"key":"ref16","article-title":"Very deep convolutional networks for large-scale image recognition","author":"Simonyan","year":"2014","journal-title":"arXiv:1409.1556"},{"doi-asserted-by":"publisher","key":"ref17","DOI":"10.5555\/3045118.3045167"},{"doi-asserted-by":"publisher","key":"ref18","DOI":"10.1109\/CVPR.2016.308"},{"doi-asserted-by":"publisher","key":"ref19","DOI":"10.1007\/978-3-030-01216-8_12"},{"key":"ref20","article-title":"CoCa: Contrastive captioners are image-text foundation models","author":"Yu","year":"2022","journal-title":"arXiv:2205.01917"},{"doi-asserted-by":"publisher","key":"ref21","DOI":"10.1109\/ACCESS.2014.2325029"},{"doi-asserted-by":"publisher","key":"ref22","DOI":"10.1109\/BigData47090.2019.9006412"},{"doi-asserted-by":"publisher","key":"ref23","DOI":"10.5555\/3454287.3455008"},{"doi-asserted-by":"publisher","key":"ref24","DOI":"10.1109\/TBDATA.2015.2472014"},{"key":"ref25","article-title":"MXNet: A flexible and efficient machine learning library for heterogeneous distributed systems","author":"Chen","year":"2015","journal-title":"arXiv:1512.01274"},{"key":"ref26","first-page":"265","article-title":"{TensorFlow}: A system for {Large\u2013Scale} machine learning","volume-title":"Proc. 12th USENIX Symp. operating Syst. design Implement. (OSDI)","author":"Abadi"},{"key":"ref27","first-page":"1","article-title":"Blueconnect: Novel hierarchical all-reduce on multi-tired network for deep learning","volume-title":"Proc. 2nd SysML Conf.","author":"Cho"},{"doi-asserted-by":"publisher","key":"ref28","DOI":"10.1145\/3448016.3452773"},{"doi-asserted-by":"publisher","key":"ref29","DOI":"10.1016\/j.jpdc.2008.09.002"},{"key":"ref30","first-page":"1","article-title":"Large scale distributed deep networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"25","author":"Dean"},{"doi-asserted-by":"publisher","key":"ref31","DOI":"10.14778\/1920841.1920931"},{"key":"ref32","first-page":"2","article-title":"Parameter server for distributed machine learning","volume-title":"Proc. Big Learn. NIPS workshop","volume":"6","author":"Li"},{"doi-asserted-by":"publisher","key":"ref33","DOI":"10.5555\/2685048.2685095"},{"key":"ref34","first-page":"172","article-title":"Blink: Fast and generic collectives for distributed ML","volume-title":"Proc. Mach. Learn. Syst.","author":"Wang"},{"doi-asserted-by":"publisher","key":"ref35","DOI":"10.1145\/3415749"},{"doi-asserted-by":"publisher","key":"ref36","DOI":"10.1109\/BigData52589.2021.9672001"},{"doi-asserted-by":"publisher","key":"ref37","DOI":"10.1109\/ICDCS51616.2021.00057"},{"doi-asserted-by":"publisher","key":"ref38","DOI":"10.1177\/1094342005051521"},{"doi-asserted-by":"publisher","key":"ref39","DOI":"10.1109\/ICDM.2016.0028"},{"year":"2009","author":"Krizhevsky","article-title":"Learning multiple layers of features from tiny images","key":"ref40"},{"doi-asserted-by":"publisher","key":"ref41","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref42","first-page":"1223","article-title":"More effective distributed ML via a stale synchronous parallel parameter server","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Ho"},{"key":"ref43","first-page":"8056","article-title":"Pipe-SGD: A decentralized pipelined SGD framework for distributed deep net training","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"31","author":"Li"},{"doi-asserted-by":"publisher","key":"ref44","DOI":"10.1109\/CVPR.2016.284"},{"doi-asserted-by":"publisher","key":"ref45","DOI":"10.1109\/MNET.011.2000530"},{"issue":"2","key":"ref46","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3377454","article-title":"A survey on distributed machine learning","volume":"53","author":"Verbraeken","year":"2020","journal-title":"ACM Comput. Surv."},{"doi-asserted-by":"publisher","key":"ref47","DOI":"10.1109\/JIOT.2024.3407584"},{"key":"ref48","first-page":"1","article-title":"Epidemic learning: Boosting decentralized learning with randomized communication","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"36","author":"Vos"},{"doi-asserted-by":"publisher","key":"ref49","DOI":"10.1109\/TBDATA.2022.3222971"},{"doi-asserted-by":"publisher","key":"ref50","DOI":"10.1109\/ACCESS.2024.3442014"},{"doi-asserted-by":"publisher","key":"ref51","DOI":"10.1145\/3373376.3378499"},{"key":"ref52","first-page":"3043","article-title":"Asynchronous decentralized parallel stochastic gradient descent","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Lian"},{"doi-asserted-by":"publisher","key":"ref53","DOI":"10.1109\/ICDCS.2019.00150"},{"doi-asserted-by":"publisher","key":"ref54","DOI":"10.1109\/TNET.2023.3328632"},{"doi-asserted-by":"publisher","key":"ref55","DOI":"10.1109\/TNET.2020.3035770"},{"doi-asserted-by":"publisher","key":"ref56","DOI":"10.1109\/TII.2022.3223234"},{"doi-asserted-by":"publisher","key":"ref57","DOI":"10.1109\/TNNLS.2022.3166101"},{"doi-asserted-by":"publisher","key":"ref58","DOI":"10.1145\/3267809.3275463"},{"doi-asserted-by":"publisher","key":"ref59","DOI":"10.1145\/3097983.3098029"},{"key":"ref60","first-page":"418","article-title":"TicTac: Accelerating distributed deep learning with communication scheduling","volume-title":"Proc. Mach. Learn. Syst.","author":"Hashemi"},{"doi-asserted-by":"publisher","key":"ref61","DOI":"10.1145\/2806777.2806778"},{"key":"ref62","first-page":"1508","article-title":"TernGrad: Ternary gradients to reduce communication in distributed deep learning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"30","author":"Wen"},{"key":"ref63","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":"ref64","first-page":"196","article-title":"Communication-efficient distributed optimization with quantized preconditioners","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Alimisis"},{"doi-asserted-by":"publisher","key":"ref65","DOI":"10.14778\/3476311.3476325"},{"doi-asserted-by":"publisher","key":"ref66","DOI":"10.1145\/3514221.3517860"}],"container-title":["IEEE Access"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/6287639\/10820123\/10836684.pdf?arnumber=10836684","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,1,20]],"date-time":"2025-01-20T19:04:31Z","timestamp":1737399871000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10836684\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025]]},"references-count":66,"URL":"https:\/\/doi.org\/10.1109\/access.2025.3528248","relation":{},"ISSN":["2169-3536"],"issn-type":[{"type":"electronic","value":"2169-3536"}],"subject":[],"published":{"date-parts":[[2025]]}}}