{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,18]],"date-time":"2025-10-18T10:52:06Z","timestamp":1760784726519,"version":"3.37.3"},"reference-count":46,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","license":[{"start":{"date-parts":[[2019,1,1]],"date-time":"2019-01-01T00:00:00Z","timestamp":1546300800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["61425016","61722211","61773362","61872338"],"award-info":[{"award-number":["61425016","61722211","61773362","61872338"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100002367","name":"Chinese Academy of Sciences","doi-asserted-by":"publisher","award":["20144310","2016102"],"award-info":[{"award-number":["20144310","2016102"]}],"id":[{"id":"10.13039\/501100002367","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100012166","name":"National Key Research and Development Program of China","doi-asserted-by":"crossref","award":["2016QY02D0405"],"award-info":[{"award-number":["2016QY02D0405"]}],"id":[{"id":"10.13039\/501100012166","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100002865","name":"Chongqing Science and Technology Commission","doi-asserted-by":"publisher","award":["cstc2017jcyjBX0059"],"award-info":[{"award-number":["cstc2017jcyjBX0059"]}],"id":[{"id":"10.13039\/501100002865","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Access"],"published-print":{"date-parts":[[2019]]},"DOI":"10.1109\/access.2019.2949611","type":"journal-article","created":{"date-parts":[[2019,10,25]],"date-time":"2019-10-25T19:56:56Z","timestamp":1572033416000},"page":"156848-156859","source":"Crossref","is-referenced-by-count":7,"title":["Trend-Smooth: Accelerate Asynchronous SGD by Smoothing Parameters Using Parameter Trends"],"prefix":"10.1109","volume":"7","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-7754-0967","authenticated-orcid":false,"given":"Guoxin","family":"Cui","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jiafeng","family":"Guo","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yixing","family":"Fan","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yanyan","family":"Lan","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xueqi","family":"Cheng","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref39","first-page":"3054","article-title":"A comprehensive linear speedup analysis for asynchronous stochastic parallel optimization from zeroth-order to first-order","author":"lian","year":"2016","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref38","article-title":"Staleness-aware async-SGD for distributed deep learning","author":"zhang","year":"2015","journal-title":"arXiv 1511 05950"},{"key":"ref33","first-page":"693","article-title":"Hogwild: A lock-free approach to parallelizing stochastic gradient descent","author":"recht","year":"2011","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref32","first-page":"685","article-title":"Deep learning with elastic averaging SGD","volume":"1","author":"zhang","year":"2015","journal-title":"Proc 28th Int Conf Neural Inf Process Syst"},{"key":"ref31","first-page":"285","article-title":"An asynchronous parallel stochastic coordinate descent algorithm","volume":"16","author":"liu","year":"2015","journal-title":"J Mach Learn Res"},{"key":"ref30","first-page":"571","article-title":"Project adam: Building an efficient and scalable deep learning training system","author":"chilimbi","year":"2014","journal-title":"Proc of USENIX Symp on Operating Systems Design and Implementation (OSDI)"},{"key":"ref37","first-page":"2737","article-title":"Asynchronous parallel stochastic gradient for nonconvex optimization","author":"lian","year":"2015","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref36","first-page":"873","article-title":"Distributed delayed stochastic optimization","author":"agarwal","year":"2011","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref35","first-page":"2915","article-title":"Delay-tolerant algorithms for asynchronous distributed online learning","author":"mcmahan","year":"2014","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1145\/2814566"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1137\/16M1080173"},{"key":"ref40","article-title":"Horovod: Fast and easy distributed deep learning in tensorflow","author":"sergeev","year":"2018","journal-title":"arXiv 1802 05799"},{"key":"ref11","first-page":"1737","article-title":"SGD-QN: Careful Quasi-Newton stochastic gradient descent","volume":"10","author":"bordes","year":"2009","journal-title":"J Mach Learn Res"},{"key":"ref12","first-page":"436","article-title":"A stochastic Quasi-Newton method for online convex optimization","author":"schraudolph","year":"2007","journal-title":"Artificial Intelligence and Statistics"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/TSP.2014.2357775"},{"key":"ref14","first-page":"3151","article-title":"Global convergence of online limited memory BFGS","volume":"16","author":"mokhtari","year":"2015","journal-title":"J Mach Learn Res"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1137\/070704277"},{"key":"ref16","first-page":"2121","article-title":"Adaptive subgradient methods for online learning and stochastic optimization","volume":"12","author":"duchi","year":"2011","journal-title":"J Mach Learn Res"},{"key":"ref17","article-title":"An overview of gradient descent optimization algorithms","author":"ruder","year":"2016","journal-title":"arXiv 1609 04747"},{"key":"ref18","first-page":"4148","article-title":"The marginal value of adaptive gradient methods in machine learning","author":"wilson","year":"2017","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref19","article-title":"Learning personalized end-to-end goal-oriented dialog","author":"luo","year":"2018","journal-title":"arXiv 1811 04604"},{"key":"ref28","doi-asserted-by":"crossref","first-page":"251","DOI":"10.1006\/jpdc.1994.1085","article-title":"Direct bulk-synchronous parallel algorithms","volume":"22","author":"gerbessiotis","year":"1994","journal-title":"J Parallel Distrib Comput"},{"key":"ref4","first-page":"3","article-title":"Scaling distributed machine learning with the parameter server","volume":"1","author":"li","year":"2014","journal-title":"Proc OSDI"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1145\/1327452.1327492"},{"key":"ref3","article-title":"Adam: A method for stochastic optimization","author":"kingma","year":"2014","journal-title":"arXiv 1412 6980"},{"key":"ref6","first-page":"1223","article-title":"Large scale distributed deep networks","volume":"1","author":"dean","year":"2012","journal-title":"Proc 25th Int Conf Neural Inf Process Syst"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1007\/978-1-4615-4123-3_4"},{"key":"ref5","first-page":"1223","article-title":"More effective distributed ML via a stale synchronous parallel parameter server","author":"ho","year":"2013","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref8","article-title":"Introspection: Accelerating neural network training by learning weight evolution","author":"sinha","year":"2017","journal-title":"arXiv 1704 04959"},{"key":"ref7","first-page":"4120","article-title":"Asynchronous stochastic gradient descent with delay compensation","volume":"70","author":"zheng","year":"2017","journal-title":"Proc 34th Int Conf Mach Learn"},{"key":"ref2","first-page":"26","article-title":"Lecture 6.5-RMSPROP: Divide the gradient by a running average of its recent magnitude","volume":"4","author":"tieleman","year":"2012","journal-title":"COURSERA Neural Netw Mach Learn"},{"key":"ref9","article-title":"Tensorflow: Large-scale machine learning on heterogeneous distributed systems","author":"abadi","year":"2015","journal-title":"arXiv 1603 04467"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1016\/S0893-6080(98)00116-6"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1145\/3065386"},{"key":"ref20","first-page":"3","article-title":"Group normalization","author":"wu","year":"2018","journal-title":"Proc Eur Conf Comput Vis (ECCV)"},{"journal-title":"MNIST Handwritten Digit Database","year":"2010","author":"lecun","key":"ref45"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1016\/0167-8191(96)00024-5"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.7551\/mitpress\/7055.001.0001"},{"key":"ref42","article-title":"Accurate, large minibatch SGD: Training imagenet in 1 hour","author":"goyal","year":"2017","journal-title":"arXiv 1706 02677"},{"journal-title":"Hadoop The Definitive Guide","year":"2012","author":"white","key":"ref24"},{"key":"ref41","article-title":"PowerAI DDL","author":"cho","year":"2017","journal-title":"arXiv 1708 02188"},{"journal-title":"Mahout in Action","year":"2012","author":"owen","key":"ref23"},{"key":"ref44","article-title":"Learning multiple layers of features from tiny images","volume":"1","author":"krizhevsky","year":"2009"},{"key":"ref26","first-page":"10","article-title":"Spark: Cluster computing with working sets","author":"zaharia","year":"2010","journal-title":"Proc 2nd USENIX Conf Hot Topics Cloud Comput"},{"key":"ref43","article-title":"Highly scalable deep learning training system with mixed-precision: Training imagenet in four minutes","author":"jia","year":"2018","journal-title":"arXiv 1807 11205"},{"key":"ref25","first-page":"1235","article-title":"Mllib: Machine learning in apache spark","volume":"17","author":"meng","year":"2016","journal-title":"J Mach Learn Res"}],"container-title":["IEEE Access"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/6287639\/8600701\/08883154.pdf?arnumber=8883154","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,1,26]],"date-time":"2022-01-26T02:29:43Z","timestamp":1643164183000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/8883154\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2019]]},"references-count":46,"URL":"https:\/\/doi.org\/10.1109\/access.2019.2949611","relation":{},"ISSN":["2169-3536"],"issn-type":[{"type":"electronic","value":"2169-3536"}],"subject":[],"published":{"date-parts":[[2019]]}}}