{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,9]],"date-time":"2026-05-09T15:15:56Z","timestamp":1778339756233,"version":"3.51.4"},"reference-count":46,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"9","license":[{"start":{"date-parts":[[2024,9,1]],"date-time":"2024-09-01T00:00:00Z","timestamp":1725148800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2024,9,1]],"date-time":"2024-09-01T00:00:00Z","timestamp":1725148800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2024,9,1]],"date-time":"2024-09-01T00:00:00Z","timestamp":1725148800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"National Key Research and Development Program of China","award":["2018YFB1800800"],"award-info":[{"award-number":["2018YFB1800800"]}]},{"name":"Basic Research"},{"name":"Guangdong Research"},{"name":"Guangdong Provincial Key Laboratory of Future Networks of Intelligence","award":["2022B1212010001"],"award-info":[{"award-number":["2022B1212010001"]}]},{"DOI":"10.13039\/501100001809","name":"NSFC","doi-asserted-by":"publisher","award":["61931024"],"award-info":[{"award-number":["61931024"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"NSFC","doi-asserted-by":"publisher","award":["81922046"],"award-info":[{"award-number":["81922046"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"NSFC-Youth","doi-asserted-by":"publisher","award":["62106154"],"award-info":[{"award-number":["62106154"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Zelixir Biotechnology Company Fund"},{"name":"Tencent Open Fund"},{"name":"Information Technology Services Office (ITSO) at The Chinese University of Hong Kong, Shenzhen (CUHKSZ), Shenzhen, China"},{"name":"Singapore Government\u2019s Research, Innovation and Enterprise 2020 Plan","award":["A1892b0026"],"award-info":[{"award-number":["A1892b0026"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Neural Netw. Learning Syst."],"published-print":{"date-parts":[[2024,9]]},"DOI":"10.1109\/tnnls.2023.3255540","type":"journal-article","created":{"date-parts":[[2023,4,27]],"date-time":"2023-04-27T18:20:49Z","timestamp":1682619649000},"page":"12276-12287","source":"Crossref","is-referenced-by-count":7,"title":["Hierarchical Weight Averaging for Deep Neural Networks"],"prefix":"10.1109","volume":"35","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7948-7672","authenticated-orcid":false,"given":"Xiaozhe","family":"Gu","sequence":"first","affiliation":[{"name":"Future Network of Intelligence Institute (FNii), School of Science and Engineering (SSE), Guangdong Provincial Key Laboratory of Future Networks of Intelligence, The Chinese University of Hong Kong, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3141-1278","authenticated-orcid":false,"given":"Zixun","family":"Zhang","sequence":"additional","affiliation":[{"name":"Future Network of Intelligence Institute (FNii), School of Science and Engineering (SSE), Guangdong Provincial Key Laboratory of Future Networks of Intelligence, The Chinese University of Hong Kong, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-7658-6622","authenticated-orcid":false,"given":"Yuncheng","family":"Jiang","sequence":"additional","affiliation":[{"name":"Future Network of Intelligence Institute (FNii), School of Science and Engineering (SSE), Guangdong Provincial Key Laboratory of Future Networks of Intelligence, The Chinese University of Hong Kong, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3415-3676","authenticated-orcid":false,"given":"Tao","family":"Luo","sequence":"additional","affiliation":[{"name":"Agency for Science, Technology and Research (A*STAR), Institute of High Performance Computing, Fusionopolis, Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9511-7532","authenticated-orcid":false,"given":"Ruimao","family":"Zhang","sequence":"additional","affiliation":[{"name":"School of Data Science (SDS), The Chinese University of Hong Kong, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9205-1881","authenticated-orcid":false,"given":"Shuguang","family":"Cui","sequence":"additional","affiliation":[{"name":"School of Science and Engineering (SSE)Guangdong Provincial Key Laboratory of Future Networks of Intelligence, Future Network of Intelligence Institute (FNii), The Chinese University of Hong Kong, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7669-2686","authenticated-orcid":false,"given":"Zhen","family":"Li","sequence":"additional","affiliation":[{"name":"Future Network of Intelligence Institute (FNii), School of Science and Engineering (SSE), Guangdong Provincial Key Laboratory of Future Networks of Intelligence, The Chinese University of Hong Kong, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","first-page":"1097","article-title":"ImageNet classification with deep convolutional neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Krizhevsky"},{"key":"ref2","article-title":"Learning multiple layers of features from tiny images","author":"Krizhevsky","year":"2009"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1097\/00003643-201406001-00333"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7298965"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.5194\/isprs-annals-IV-1-W1-91-2017"},{"issue":"7","key":"ref6","first-page":"1","article-title":"Adaptive subgradient methods for online learning and stochastic optimization","volume":"12","author":"Duchi","year":"2011","journal-title":"J. Mach. Learn. Res."},{"key":"ref7","article-title":"SGDR: Stochastic gradient descent with warm restarts","author":"Loshchilov","year":"2016","journal-title":"arXiv:1608.03983"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/WACV.2017.58"},{"key":"ref9","article-title":"Don\u2019t use large mini-batches, use local SGD","author":"Lin","year":"2018","journal-title":"arXiv:1808.07217"},{"key":"ref10","first-page":"1273","article-title":"Communication-efficient learning of deep networks from decentralized data","volume-title":"Proc. 20th Int. Conf. Artif. Intell. Statist.","author":"McMahan"},{"key":"ref11","first-page":"1231","article-title":"Efficient large-scale distributed training of conditional maximum entropy models","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Mcdonald"},{"key":"ref12","article-title":"Parallel SGD: When does averaging help?","author":"Zhang","year":"2016","journal-title":"arXiv:1606.07365"},{"key":"ref13","article-title":"Local SGD converges fast and communicates little","author":"Stich","year":"2018","journal-title":"arXiv:1805.09767"},{"key":"ref14","first-page":"2595","article-title":"Parallelized stochastic gradient descent","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Zinkevich"},{"key":"ref15","article-title":"Averaging weights leads to wider optima and better generalization","author":"Izmailov","year":"2018","journal-title":"arXiv:1803.05407"},{"key":"ref16","first-page":"19","article-title":"Communication efficient distributed machine learning with the parameter server","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Li"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v33i01.33015693"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref19","volume-title":"The Caltech-UCSD birds-200\u20132011 dataset","author":"Wah","year":"2011"},{"key":"ref20","volume-title":"Caltech-256 object category dataset","author":"Griffin","year":"2007"},{"key":"ref21","volume-title":"Efficient estimations from a slowly convergent Robbins\u2013Monro process","author":"Ruppert","year":"1988"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1137\/0330046"},{"key":"ref23","article-title":"Snapshot ensembles: Train 1, get m for free","author":"Huang","year":"2017","journal-title":"arXiv:1704.00109"},{"key":"ref24","first-page":"13153","article-title":"A simple baseline for Bayesian uncertainty in deep learning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Maddox"},{"key":"ref25","article-title":"SWALP: Stochastic weight averaging in low-precision training","author":"Yang","year":"2019","journal-title":"arXiv:1904.11943"},{"key":"ref26","article-title":"Stochastic weight averaging in parallel: Large-batch training that generalizes well","author":"Gupta","year":"2020","journal-title":"arXiv:2001.02312"},{"key":"ref27","first-page":"22405","article-title":"SWAD: Domain generalization by seeking flat minima","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"34","author":"Cha"},{"key":"ref28","article-title":"Diverse weight averaging for out-of-distribution generalization","author":"Ram\u00e9","year":"2022","journal-title":"arXiv:2205.09739"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/LSP.2021.3082036"},{"key":"ref30","first-page":"165","article-title":"Optimal distributed online prediction using mini-batches","volume":"13","author":"Dekel","year":"2012","journal-title":"J. Mach. Learn. Res."},{"key":"ref31","article-title":"On the convergence properties of a K-step averaging stochastic gradient descent algorithm for nonconvex optimization","author":"Zhou","year":"2017","journal-title":"arXiv:1708.01012"},{"key":"ref32","first-page":"9597","article-title":"Lookahead optimizer: k steps forward, 1 step back","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Zhang"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.01092"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00663"},{"key":"ref35","article-title":"Sharpness-aware minimization for efficiently improving generalization","author":"Foret","year":"2020","journal-title":"arXiv:2010.01412"},{"key":"ref36","first-page":"192","article-title":"The loss surfaces of multilayer networks","volume-title":"Proc. 18th Int. Conf. Artif. Intell. Statist.","author":"Choromanska"},{"key":"ref37","article-title":"Qualitatively characterizing neural network optimization problems","author":"Goodfellow","year":"2014","journal-title":"arXiv:1412.6544"},{"key":"ref38","first-page":"6389","article-title":"Visualizing the loss landscape of neural nets","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Li"},{"key":"ref39","article-title":"The loss landscape of overparameterized neural networks","author":"Cooper","year":"2018","journal-title":"arXiv:1804.10200"},{"key":"ref40","first-page":"1309","article-title":"Essentially no barriers in neural network energy landscape","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Draxler"},{"key":"ref41","article-title":"On large-batch training for deep learning: Generalization gap and sharp minima","author":"Keskar","year":"2016","journal-title":"arXiv:1609.04836"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00381"},{"key":"ref43","article-title":"Asymmetric valleys: Beyond sharp and flat local minima","author":"He","year":"2019","journal-title":"arXiv:1902.00744"},{"issue":"5","key":"ref44","first-page":"8","article-title":"Tiny ImageNet classification with convolutional neural networks","volume":"2","author":"Yao","year":"2015","journal-title":"CS 231N"},{"key":"ref45","volume-title":"Code of Filter Grafting","author":"Meng","year":"2020"},{"key":"ref46","article-title":"Improved regularization of convolutional neural networks with cutout","author":"DeVries","year":"2017","journal-title":"arXiv:1708.04552"}],"container-title":["IEEE Transactions on Neural Networks and Learning Systems"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/5962385\/10663876\/10109817.pdf?arnumber=10109817","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,9,4]],"date-time":"2024-09-04T12:19:15Z","timestamp":1725452355000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10109817\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,9]]},"references-count":46,"journal-issue":{"issue":"9"},"URL":"https:\/\/doi.org\/10.1109\/tnnls.2023.3255540","relation":{},"ISSN":["2162-237X","2162-2388"],"issn-type":[{"value":"2162-237X","type":"print"},{"value":"2162-2388","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,9]]}}}