{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,30]],"date-time":"2026-05-30T01:26:41Z","timestamp":1780104401591,"version":"3.54.0"},"reference-count":54,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","license":[{"start":{"date-parts":[[2023,1,1]],"date-time":"2023-01-01T00:00:00Z","timestamp":1672531200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2023,1,1]],"date-time":"2023-01-01T00:00:00Z","timestamp":1672531200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2023,1,1]],"date-time":"2023-01-01T00:00:00Z","timestamp":1672531200000},"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":["IEEE Trans. Signal Process."],"published-print":{"date-parts":[[2023]]},"DOI":"10.1109\/tsp.2023.3309461","type":"journal-article","created":{"date-parts":[[2023,9,1]],"date-time":"2023-09-01T17:37:19Z","timestamp":1693589839000},"page":"3257-3266","source":"Crossref","is-referenced-by-count":24,"title":["Efficient-Adam: Communication-Efficient Distributed Adam"],"prefix":"10.1109","volume":"71","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-9795-4200","authenticated-orcid":false,"given":"Congliang","family":"Chen","sequence":"first","affiliation":[{"name":"Chinese University of Hong Kong (Shenzhen), Shenzhen, Guangdong, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5659-3464","authenticated-orcid":false,"given":"Li","family":"Shen","sequence":"additional","affiliation":[{"name":"JD Explore Academy, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3865-8145","authenticated-orcid":false,"given":"Wei","family":"Liu","sequence":"additional","affiliation":[{"name":"Tencent, Shenzhen, Guangdong, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3995-914X","authenticated-orcid":false,"given":"Zhi-Quan","family":"Luo","sequence":"additional","affiliation":[{"name":"Chinese University of Hong Kong (Shenzhen), Shenzhen, Guangdong, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref13","first-page":"1","article-title":"Analysis of quantized models","author":"hou","year":"2019","journal-title":"Proc Int Conf Learn Represent"},{"key":"ref12","first-page":"1","article-title":"Large batch optimization for deep learning: Training BERT in 76 minutes","author":"you","year":"0","journal-title":"Proc Int Conf Learn Represent"},{"key":"ref15","first-page":"8024","article-title":"PyTorch: An imperative style, high-performance deep learning library","author":"paszke","year":"0","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref14","article-title":"Adam: A method for stochastic optimization","author":"kingma","year":"0"},{"key":"ref53","first-page":"142","article-title":"Learning word vectors for sentiment analysis","author":"maas","year":"2011","journal-title":"Proc 49th Annu Meet Assoc Comput Ling Hum Lang Technol"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.3115\/v1\/W14-4012"},{"key":"ref11","article-title":"Accurate, large minibatch SGD: Training ImageNet in 1 hour","author":"goyal","year":"0"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1145\/2623330.2623612"},{"key":"ref54","article-title":"BERT: Pre-training of deep bidirectional transformers for language understanding","author":"devlin","year":"0"},{"key":"ref17","article-title":"MXNet: A flexible and efficient machine learning library for heterogeneous distributed systems","author":"chen","year":"0"},{"key":"ref16","first-page":"265","article-title":"TensorFlow: A system for large-scale machine learning","author":"abadi","year":"0","journal-title":"Proc 12th USENIX Symp Operating Syst Des Implementation (OSDI)"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1145\/2640087.2644155"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.14778\/1920841.1920931"},{"key":"ref51","article-title":"The marginal value of adaptive gradient methods in machine learning","author":"wilson","year":"0"},{"key":"ref50","article-title":"Learning multiple layers of features from tiny images","author":"krizhevsky","year":"2009"},{"key":"ref46","first-page":"1","article-title":"RMSprop converges with proper hyper-parameter","author":"shi","year":"2021","journal-title":"Proc Int Conf Learn Represent"},{"key":"ref45","article-title":"AdaSAM: Boosting sharpness-aware minimization with adaptive learning rate and momentum for training deep neural networks","author":"sun","year":"0"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/TSP.2022.3223214"},{"key":"ref47","article-title":"Adam can converge without any modification on update rules","author":"zhang","year":"0"},{"key":"ref42","article-title":"On the convergence of Adam and Adagrad","author":"d\u00e9fossez","year":"0"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2023.3279381"},{"key":"ref44","article-title":"On the convergence of a class of Adam-type algorithms for non-convex optimization","author":"chen","year":"0"},{"key":"ref43","article-title":"On the convergence of Adam and beyond","author":"reddi","year":"0"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/W18-5446"},{"key":"ref7","first-page":"740","article-title":"Microsoft COCO: Common objects in context","author":"lin","year":"0","journal-title":"Proc Eur Conf Comput Vis"},{"key":"ref9","article-title":"On efficient training of large-scale deep learning models: A literature review","author":"shen","year":"0"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1201\/b18401"},{"key":"ref3","author":"sutton","year":"2018","journal-title":"Reinforcement Learning An Introduction"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2013.50"},{"key":"ref40","article-title":"AdaGrad stepsizes: Sharp convergence over nonconvex landscapes, from any initialization","author":"ward","year":"0"},{"key":"ref35","article-title":"DeepSqueeze: Parallel stochastic gradient descent with double-pass error-compensated compression","author":"tang","year":"0"},{"key":"ref34","article-title":"Decentralized stochastic optimization and gossip algorithms with compressed communication","author":"koloskova","year":"0"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/OJCSYS.2022.3210453"},{"key":"ref36","first-page":"6292","article-title":"Communication-compressed adaptive gradient method for distributed nonconvex optimization","author":"wang","year":"0","journal-title":"Proc Int Conf Artif Intell Statist"},{"key":"ref31","first-page":"2525","article-title":"A linear speedup analysis of distributed deep learning with sparse and quantized communication","author":"jiang","year":"0","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref30","article-title":"signSGD: Compressed optimisation for non-convex problems","author":"bernstein","year":"0"},{"key":"ref33","first-page":"1509","article-title":"TernGrad: Ternary gradients to reduce communication in distributed deep learning","author":"wen","year":"0","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref32","first-page":"1299","article-title":"Gradient sparsification for communication-efficient distributed optimization","author":"wangni","year":"0","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref2","author":"goodfellow","year":"2016","journal-title":"Deep Learning"},{"key":"ref1","author":"bishop","year":"2006","journal-title":"Pattern Recognition and Machine Learning"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/JSTSP.2018.2848718"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.23919\/ECC55457.2022.9838141"},{"key":"ref24","first-page":"10411","article-title":"Towards practical Adam: Non-convexity, convergence theory, and mini-batch acceleration","volume":"23","author":"chen","year":"0","journal-title":"J Mach Learn Res"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.01138"},{"key":"ref26","article-title":"Error feedback fixes signSGD and other gradient compression schemes","author":"karimireddy","year":"0"},{"key":"ref25","first-page":"1709","article-title":"QSGD: Communication-efficient SGD via gradient quantization and encoding","author":"alistarh","year":"0","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref20","article-title":"The error-feedback framework: Better rates for SGD with delayed gradients and compressed communication","author":"stich","year":"0"},{"key":"ref22","first-page":"11446","article-title":"Communication-efficient distributed blockwise momentum SGD with error-feedback","author":"zheng","year":"0","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref21","first-page":"4447","article-title":"Sparsified SGD with memory","author":"stich","year":"0","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/CDC.2012.6426626"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1145\/3470890"},{"key":"ref29","first-page":"5330","article-title":"Can decentralized algorithms outperform centralized algorithms? A case study for decentralized parallel stochastic gradient descent","author":"lian","year":"0","journal-title":"Proc Adv Neural Inf Process Syst"}],"container-title":["IEEE Transactions on Signal Processing"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/78\/10040758\/10237319.pdf?arnumber=10237319","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,10,16]],"date-time":"2023-10-16T18:33:33Z","timestamp":1697481213000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10237319\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023]]},"references-count":54,"URL":"https:\/\/doi.org\/10.1109\/tsp.2023.3309461","relation":{},"ISSN":["1053-587X","1941-0476"],"issn-type":[{"value":"1053-587X","type":"print"},{"value":"1941-0476","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023]]}}}