{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,7]],"date-time":"2026-09-07T20:09:42Z","timestamp":1788811782339,"version":"build-2803163510"},"reference-count":34,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"3","license":[{"start":{"date-parts":[[2026,7,1]],"date-time":"2026-07-01T00:00:00Z","timestamp":1782864000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,7,1]],"date-time":"2026-07-01T00:00:00Z","timestamp":1782864000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,7,1]],"date-time":"2026-07-01T00:00:00Z","timestamp":1782864000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"Taishan Industrial Experts Program"},{"DOI":"10.13039\/501100012166","name":"National Key R&D Program of China","doi-asserted-by":"crossref","award":["2024YFE0210800"],"award-info":[{"award-number":["2024YFE0210800"]}],"id":[{"id":"10.13039\/501100012166","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Cloud Comput."],"published-print":{"date-parts":[[2026,7]]},"DOI":"10.1109\/tcc.2026.3706801","type":"journal-article","created":{"date-parts":[[2026,6,24]],"date-time":"2026-06-24T19:50:58Z","timestamp":1782330658000},"page":"1975-1988","source":"Crossref","is-referenced-by-count":0,"title":["Mixer: Accelerating Deep Neural Network Training With Hybrid Optimizers"],"prefix":"10.1109","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-4825-5673","authenticated-orcid":false,"given":"Ruidong","family":"Yan","sequence":"first","affiliation":[{"name":"IEIT SYSTEMS Company, Ltd.","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7656-6428","authenticated-orcid":false,"given":"Jidong","family":"Zhai","sequence":"additional","affiliation":[{"name":"Tsinghua University","place":["Beijing, China"],"department":["Department of Computer Science and Technology"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-4460-9869","authenticated-orcid":false,"given":"Rengang","family":"Li","sequence":"additional","affiliation":[{"name":"IEIT SYSTEMS Company, Ltd.","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhiyong","family":"Qiu","sequence":"additional","affiliation":[{"name":"IEIT SYSTEMS Company, Ltd.","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Liang","family":"Jin","sequence":"additional","affiliation":[{"name":"IEIT SYSTEMS Company, Ltd.","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2358-3395","authenticated-orcid":false,"given":"Yuyang","family":"Jin","sequence":"additional","affiliation":[{"name":"Tsinghua University","place":["Beijing, China"],"department":["Department of Computer Science and Technology"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yaqian","family":"Zhao","sequence":"additional","affiliation":[{"name":"IEIT SYSTEMS Company, Ltd.","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1303-6681","authenticated-orcid":false,"given":"Zhenhua","family":"Guo","sequence":"additional","affiliation":[{"name":"IEIT SYSTEMS Company, Ltd.","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","first-page":"1","article-title":"Large batch optimization for deep learning: Training BERT in 76 minutes","volume-title":"Proc. 8th Int. Conf. Learn. Representations","author":"You"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/hipc56025.2022.00044"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1137\/16M1080173"},{"key":"ref4","article-title":"Adam: A method for stochastic optimization","author":"Kingma","year":"2014"},{"key":"ref5","first-page":"2408","article-title":"Optimizing neural networks with Kronecker-factored approximate curvature","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Martens"},{"key":"ref6","article-title":"Scalable second order optimization for deep learning","author":"Anil","year":"2020"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2024.3423382"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1145\/3394486.3403265"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejor.2005.06.076"},{"key":"ref10","first-page":"8082","article-title":"Fast convergence of natural gradient descent for overparameterized neural networks","volume-title":"Proc. 33rd Int. Conf. Neural Inf. Process. Syst.","author":"Zhang"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2022.3213654"},{"key":"ref12","article-title":"Eva: A general vectorized approximation framework for second-order optimization","author":"Zhang","year":"2023"},{"key":"ref13","first-page":"1","article-title":"KAISA: An adaptive second-order optimizer framework for deep neural networks","volume-title":"Proc. Int. Conf. High Perform. Comput. Netw. Storage Anal.","author":"Pauloski"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.3390\/ijgi8090366"},{"key":"ref15","article-title":"Extremely large minibatch SGD: Training ResNet-50 on ImageNet in 15 minutes","author":"Akiba","year":"2017"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v35i8.16867"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v35i9.16921"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1201\/9781003162810-13"},{"key":"ref20","article-title":"Making covariance matricespositive definite","author":"Schaeffer","year":"2017"},{"key":"ref21","article-title":"Wikipedia corpus","year":"2021"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.11"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D16-1264"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/n19-1423"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.01264"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/SC41405.2020.00098"},{"key":"ref27","article-title":"TENGraD: Time-efficient natural gradient descent with exact fisher-block inversion","author":"Soori","year":"2021"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/SC41404.2022.00052"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2020.3004354"},{"key":"ref30","article-title":"Improving generalization performance by switching from ADAM to SGD","author":"Keskar","year":"2017"},{"key":"ref31","article-title":"FOSI: Hybrid first and second order optimization","author":"Sivan","year":"2023"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.6028\/jres.045.026"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1007\/s00778-020-00636-3"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1145\/3710848.3710852"}],"container-title":["IEEE Transactions on Cloud Computing"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/6245519\/11678135\/11576622.pdf?arnumber=11576622","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,7]],"date-time":"2026-09-07T19:16:16Z","timestamp":1788808576000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11576622\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7]]},"references-count":34,"journal-issue":{"issue":"3"},"URL":"https:\/\/doi.org\/10.1109\/tcc.2026.3706801","relation":{},"ISSN":["2168-7161"],"issn-type":[{"value":"2168-7161","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7]]}}}