{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,31]],"date-time":"2026-01-31T10:14:04Z","timestamp":1769854444952,"version":"3.49.0"},"reference-count":31,"publisher":"IEEE","license":[{"start":{"date-parts":[[2021,6,6]],"date-time":"2021-06-06T00:00:00Z","timestamp":1622937600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2021,6,6]],"date-time":"2021-06-06T00:00:00Z","timestamp":1622937600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2021,6,6]]},"DOI":"10.1109\/icassp39728.2021.9414564","type":"proceedings-article","created":{"date-parts":[[2021,5,13]],"date-time":"2021-05-13T19:53:45Z","timestamp":1620935625000},"page":"3675-3679","source":"Crossref","is-referenced-by-count":11,"title":["Decentralized Deep Learning Using Momentum-Accelerated Consensus"],"prefix":"10.1109","author":[{"given":"Aditya","family":"Balu","sequence":"first","affiliation":[{"name":"Iowa State University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhanhong","family":"Jiang","sequence":"additional","affiliation":[{"name":"Johnson Controls"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Sin Yong","family":"Tan","sequence":"additional","affiliation":[{"name":"Iowa State University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Chinmay","family":"Hedge","sequence":"additional","affiliation":[{"name":"New York University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Young M","family":"Lee","sequence":"additional","affiliation":[{"name":"Johnson Controls"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Soumik","family":"Sarkar","sequence":"additional","affiliation":[{"name":"Iowa State University"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref31","article-title":"Communication efficient decentralized training with multiple local updates","author":"li","year":"2019","journal-title":"arXiv preprint arXiv 1910 09643"},{"key":"ref30","article-title":"Snapshot ensembles: Train 1, get m for free","author":"huang","year":"2017","journal-title":"arXiv preprint arXiv 1704 00109"},{"key":"ref10","first-page":"5336","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":"ref11","first-page":"5906","article-title":"Collaborative deep learning in fixed topology networks","author":"jiang","year":"2017","journal-title":"Advances in neural information processing systems"},{"key":"ref12","article-title":"Stochastic gradient push for distributed deep learning","author":"assran","year":"2018","journal-title":"arXiv preprint arXiv 1811 10792"},{"key":"ref13","first-page":"393","article-title":"Efficient decentralized deep learning by dynamic model averaging","author":"kamp","year":"2018","journal-title":"Proceedings of the European Conference on Machine Learning and Knowledge Discovery in Databases"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1016\/j.dsp.2018.03.006"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/SAM.2016.7569731"},{"key":"ref16","article-title":"Parallelizing mcmc via weierstrass sampler","author":"wang","year":"2013","journal-title":"arXiv preprint arXiv 1312 4605"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/TSP.2013.2278149"},{"key":"ref18","article-title":"Linearly convergent stochastic heavy ball method for minimizing generalization error","author":"loizou","year":"2017","journal-title":"arXiv preprint arXiv 1710 10737"},{"key":"ref19","article-title":"Unified convergence analysis of stochastic momentum methods for convex and non-convex optimization","author":"yang","year":"2016","journal-title":"arXiv preprint arXiv 1604 03257"},{"key":"ref28","article-title":"D2: Decentralized training over decentralized data","author":"tang","year":"2018","journal-title":"arXiv preprint arXiv 1803 07068"},{"key":"ref4","article-title":"Advances and open problems in federated learning","author":"kairouz","year":"2019","journal-title":"arXiv preprint arXiv 1912 04977"},{"key":"ref27","first-page":"3059","article-title":"An accelerated proximal coordinate gradient method","author":"lin","year":"2014","journal-title":"Advances in neural information processing systems"},{"key":"ref3","article-title":"How to scale distributed deep learning?","author":"jin","year":"2016","journal-title":"arXiv preprint arXiv 1611 04581"},{"key":"ref6","article-title":"Accurate, large minibatch sgd: Training imagenet in 1 hour","author":"goyal","year":"2017","journal-title":"arXiv preprint arXiv 1706 02677"},{"key":"ref29","article-title":"Averaging weights leads to wider optima and better generalization","author":"izmailov","year":"2018","journal-title":"arXiv preprint arXiv 1803 05407"},{"key":"ref5","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":"ref8","article-title":"signsgd: Compressed optimisation for non-convex problems","author":"bernstein","year":"2018","journal-title":"arXiv preprint arXiv 1802 04102"},{"key":"ref7","article-title":"Revisiting distributed synchronous sgd","author":"chen","year":"2016","journal-title":"arXiv preprint arXiv 1604 00981"},{"key":"ref2","first-page":"685","article-title":"Deep learning with elastic averaging sgd","author":"zhang","year":"2015","journal-title":"Advances in neural information processing systems"},{"key":"ref9","article-title":"Federated learning: Strategies for improving communication efficiency","author":"kone?ny","year":"2016","journal-title":"arXiv preprint arXiv 1610 05492"},{"key":"ref1","first-page":"1223","article-title":"Large scale distributed deep networks","author":"dean","year":"2012","journal-title":"Advances in neural information processing systems"},{"key":"ref20","article-title":"On the linear speedup analysis of communication efficient momentum sgd for distributed non-convex optimization","author":"yu","year":"2019","journal-title":"arXiv preprint arXiv 1905 03817"},{"key":"ref22","volume":"87","author":"nesterov","year":"2013","journal-title":"Introductory Lectures on Convex Optimization A Basic Course"},{"key":"ref21","article-title":"Slowmo: Improving communication-efficient distributed sgd with slow momentum","author":"wang","year":"2019","journal-title":"arXiv preprint arXiv 1910 12731"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1016\/0041-5553(64)90137-5"},{"key":"ref23","first-page":"1139","article-title":"On the importance of initialization and momentum in deep learning","author":"sutskever","year":"2013","journal-title":"International Conference on Machine Learning"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/TAC.2018.2880407"},{"key":"ref25","article-title":"Decentralized deep learning using momentum-accelerated consensus","author":"balu","year":"2020","journal-title":"arXiv preprint arXiv 2010 11419"}],"event":{"name":"ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","location":"Toronto, ON, Canada","start":{"date-parts":[[2021,6,6]]},"end":{"date-parts":[[2021,6,11]]}},"container-title":["ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/9413349\/9413350\/09414564.pdf?arnumber=9414564","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,8,3]],"date-time":"2022-08-03T00:19:24Z","timestamp":1659485964000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9414564\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,6,6]]},"references-count":31,"URL":"https:\/\/doi.org\/10.1109\/icassp39728.2021.9414564","relation":{},"subject":[],"published":{"date-parts":[[2021,6,6]]}}}