{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,17]],"date-time":"2026-04-17T07:57:09Z","timestamp":1776412629894,"version":"3.51.2"},"reference-count":40,"publisher":"Oxford University Press (OUP)","issue":"4","license":[{"start":{"date-parts":[[2023,6,10]],"date-time":"2023-06-10T00:00:00Z","timestamp":1686355200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/pages\/standard-publication-reuse-rights"}],"funder":[{"name":"Shanghai Research Center for Data Science and Decision Technology"},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["12071477"],"award-info":[{"award-number":["12071477"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["11831008"],"award-info":[{"award-number":["11831008"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100004260","name":"Renmin University of China","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100004260","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2023,9,29]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Distributed computing is critically important for modern statistical analysis. Herein, we develop a distributed quasi-Newton (DQN) framework with excellent statistical, computation, and communication efficiency. In the DQN method, no Hessian matrix inversion or communication is needed. This considerably reduces the computation and communication complexity of the proposed method. Notably, related existing methods only analyse numerical convergence and require a diverging number of iterations to converge. However, we investigate the statistical properties of the DQN method and theoretically demonstrate that the resulting estimator is statistically efficient over a small number of iterations under mild conditions. Extensive numerical analyses demonstrate the finite sample performance.<\/jats:p>","DOI":"10.1093\/jrsssb\/qkad059","type":"journal-article","created":{"date-parts":[[2023,6,10]],"date-time":"2023-06-10T14:15:50Z","timestamp":1686406550000},"page":"1326-1354","source":"Crossref","is-referenced-by-count":12,"title":["Quasi-Newton updating for large-scale distributed learning"],"prefix":"10.1093","volume":"85","author":[{"given":"Shuyuan","family":"Wu","sequence":"first","affiliation":[{"name":"School of Statistics and Management, Shanghai University of Finance and Economics , Shanghai , China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Danyang","family":"Huang","sequence":"additional","affiliation":[{"name":"Center for Applied Statistics and School of Statistics, Renmin University of China , Beijing , China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hansheng","family":"Wang","sequence":"additional","affiliation":[{"name":"Guanghua School of Management, Peking University , Beijing , China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"286","published-online":{"date-parts":[[2023,6,10]]},"reference":[{"issue":"3","key":"2023092923225459000_qkad059-B1","doi-asserted-by":"crossref","first-page":"223","DOI":"10.1093\/imamat\/12.3.223","article-title":"On the local and superlinear convergence of quasi-Newton methods","volume":"12","author":"Broyden","year":"1973","journal-title":"IMA Journal of Applied Mathematics"},{"issue":"3\u20134","key":"2023092923225459000_qkad059-B2","doi-asserted-by":"crossref","first-page":"231","DOI":"10.1561\/2200000050","article-title":"Theory of convex optimization for machine learning","volume":"8","author":"Bubeck","year":"2015","journal-title":"Foundations and Trends in Machine Learning"},{"key":"2023092923225459000_qkad059-B3","volume-title":"Numerical analysis","author":"Burden","year":"2015"},{"key":"2023092923225459000_qkad059-B4","article-title":"Large-scale L-BFGS using MapReduce","volume":"27","author":"Chen","year":"2014","journal-title":"Advances in Neural Information Processing Systems"},{"issue":"1\u20133","key":"2023092923225459000_qkad059-B5","doi-asserted-by":"crossref","first-page":"177","DOI":"10.1007\/BF01594934","article-title":"Convergence of quasi-Newton matrices generated by the symmetric rank one update","volume":"50","author":"Conn","year":"1991","journal-title":"Mathematical Programming"},{"key":"2023092923225459000_qkad059-B6","article-title":"DINGO: Distributed Newton-type method for gradient-norm optimization","volume":"32","author":"Crane","year":"2019","journal-title":"Advances in Neural Information Processing Systems"},{"issue":"1","key":"2023092923225459000_qkad059-B7","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1137\/0801001","article-title":"Variable metric method for minimization","volume":"1","author":"Davidon","year":"1991","journal-title":"SIAM Journal on Optimization"},{"issue":"10","key":"2023092923225459000_qkad059-B8","doi-asserted-by":"crossref","first-page":"2613","DOI":"10.1109\/TSP.2017.2666776","article-title":"Decentralized quasi-Newton methods","volume":"65","author":"Eisen","year":"2017","journal-title":"IEEE Transactions on Signal Processing"},{"issue":"456","key":"2023092923225459000_qkad059-B9","doi-asserted-by":"crossref","first-page":"1348","DOI":"10.1198\/016214501753382273","article-title":"Variable selection via nonconcave penalized likelihood and its oracle properties","volume":"96","author":"Fan","year":"2001","journal-title":"Journal of the American statistical Association"},{"issue":"5","key":"2023092923225459000_qkad059-B10","doi-asserted-by":"crossref","first-page":"849","DOI":"10.1111\/j.1467-9868.2008.00674.x","article-title":"Sure independence screening for ultra-high dimensional feature space (with discussion)","volume":"70","author":"Fan","year":"2008","journal-title":"Journal of the Royal Statistical Society, Series B"},{"issue":"6","key":"2023092923225459000_qkad059-B11","doi-asserted-by":"crossref","first-page":"3567","DOI":"10.1214\/10-AOS798","article-title":"Sure independent screening in generalized linear models with NP-dimensionality","volume":"38","author":"Fan","year":"2010","journal-title":"Annals of Statistics"},{"issue":"6","key":"2023092923225459000_qkad059-B12","doi-asserted-by":"crossref","first-page":"3009","DOI":"10.1214\/18-AOS1713","article-title":"Distributed estimation of principal eigenspaces","volume":"47","author":"Fan","year":"2019","journal-title":"Annals of Statistics"},{"issue":"109","key":"2023092923225459000_qkad059-B13","doi-asserted-by":"crossref","first-page":"23","DOI":"10.1090\/S0025-5718-1970-0258249-6","article-title":"A family of variable-metric methods derived by variational means","volume":"24","author":"Goldfarb","year":"1970","journal-title":"Mathematics of Computation"},{"key":"2023092923225459000_qkad059-B14","author":"Gopal","year":"2013"},{"key":"2023092923225459000_qkad059-B15","author":"Goyal","year":"2017"},{"issue":"1","key":"2023092923225459000_qkad059-B16","doi-asserted-by":"crossref","first-page":"342","DOI":"10.1214\/13-AOS1087","article-title":"Quantile-adaptive model-free variable screening for high-dimensional heterogeneous data","volume":"41","author":"He","year":"2013","journal-title":"Annals of Statistics"},{"issue":"173","key":"2023092923225459000_qkad059-B17","first-page":"173-1","article-title":"Doubly distributed supervised learning and inference with high-dimensional correlated outcomes","volume":"21","author":"Hector","year":"2020","journal-title":"Journal of Machine Learning Research"},{"issue":"534","key":"2023092923225459000_qkad059-B18","doi-asserted-by":"crossref","first-page":"805","DOI":"10.1080\/01621459.2020.1736082","article-title":"A distributed and integrated method of moments for high-dimensional correlated data analysis","volume":"116","author":"Hector","year":"2021","journal-title":"Journal of the American Statistical Association"},{"issue":"1\u20132","key":"2023092923225459000_qkad059-B19","doi-asserted-by":"crossref","first-page":"41","DOI":"10.1007\/s10107-019-01369-0","article-title":"A distributed one-step estimator","volume":"174","author":"Huang","year":"2019","journal-title":"Mathematical Programming"},{"issue":"526","key":"2023092923225459000_qkad059-B20","doi-asserted-by":"crossref","first-page":"668","DOI":"10.1080\/01621459.2018.1429274","article-title":"Communication-efficient distributed statistical inference","volume":"114","author":"Jordan","year":"2019","journal-title":"Journal of the American Statistical Association"},{"key":"2023092923225459000_qkad059-B21","author":"Lee","year":"2018"},{"issue":"3","key":"2023092923225459000_qkad059-B22","doi-asserted-by":"crossref","first-page":"1846","DOI":"10.1214\/12-AOS1024","article-title":"Robust rank correlation based screening","volume":"40","author":"Li","year":"2012","journal-title":"Annals of Statistics"},{"issue":"1","key":"2023092923225459000_qkad059-B23","first-page":"1","article-title":"Distributed feature screening via componentwise debiasing","volume":"21","author":"Li","year":"2020","journal-title":"Journal of Machine Learning Research"},{"issue":"2","key":"2023092923225459000_qkad059-B24","doi-asserted-by":"crossref","first-page":"249","DOI":"10.1007\/s00365-017-9379-1","article-title":"Distributed kernel-based gradient descent algorithms","volume":"47","author":"Lin","year":"2018","journal-title":"Constructive Approximation"},{"key":"2023092923225459000_qkad059-B25","author":"Mcdonald","year":"2009"},{"issue":"2","key":"2023092923225459000_qkad059-B26","doi-asserted-by":"crossref","first-page":"1670","DOI":"10.1137\/17M1122943","article-title":"IQN: An incremental quasi-Newton method with local superlinear convergence rate","volume":"28","author":"Mokhtari","year":"2018","journal-title":"SIAM Journal on Optimization"},{"key":"2023092923225459000_qkad059-B27","doi-asserted-by":"crossref","DOI":"10.1007\/b98874","volume-title":"Numerical optimization","author":"Nocedal","year":"1999"},{"issue":"6","key":"2023092923225459000_qkad059-B28","doi-asserted-by":"crossref","first-page":"2566","DOI":"10.1109\/TAC.2019.2937496","article-title":"Accelerated distributed Nesterov gradient descent","volume":"65","author":"Qu","year":"2019","journal-title":"IEEE Transactions on Automatic Control"},{"issue":"2","key":"2023092923225459000_qkad059-B29","doi-asserted-by":"crossref","first-page":"181","DOI":"10.1007\/BF01459951","article-title":"On the order of convergence of certain quasi-Newton-methods","volume":"23","author":"Schuller","year":"1974","journal-title":"Numerische Mathematik"},{"key":"2023092923225459000_qkad059-B30","author":"Shamir","year":"2014"},{"key":"2023092923225459000_qkad059-B31","doi-asserted-by":"crossref","DOI":"10.1007\/b97553","volume-title":"Mathematical statistics","author":"Shao","year":"2003"},{"key":"2023092923225459000_qkad059-B32","author":"Soori","year":"2020"},{"issue":"1","key":"2023092923225459000_qkad059-B33","first-page":"1","article-title":"Securing distributed gradient descent in high dimensional statistical learning","volume":"3","author":"Su","year":"2019","journal-title":"Proceedings of the ACM on Measurement and Analysis of Computing Systems"},{"key":"2023092923225459000_qkad059-B34","doi-asserted-by":"crossref","first-page":"104567","DOI":"10.1016\/j.jmva.2019.104567","article-title":"Distributed simultaneous inference in generalized linear models via confidence distribution","volume":"176","author":"Tang","year":"2020","journal-title":"Journal of Multivariate Analysis"},{"key":"2023092923225459000_qkad059-B35","volume-title":"Asymptotic statistics","author":"Van der Vaart","year":"2000"},{"key":"2023092923225459000_qkad059-B36","author":"Zhu","year":"2021"},{"key":"2023092923225459000_qkad059-B37","first-page":"2338","article-title":"Giant: Globally improved approximate newton method for distributed optimization","volume":"31","author":"Wang","year":"2018","journal-title":"Advances in Neural Information Processing Systems"},{"issue":"1","key":"2023092923225459000_qkad059-B38","first-page":"3321","article-title":"Communication-efficient algorithms for statistical optimization","volume":"14","author":"Zhang","year":"2013","journal-title":"The Journal of Machine Learning Research"},{"key":"2023092923225459000_qkad059-B39","author":"Zhang","year":"2015"},{"issue":"4","key":"2023092923225459000_qkad059-B40","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1080\/10618600.2021.1923517","article-title":"Least-square approximation for a distributed system","volume":"30","author":"Zhu","year":"2021","journal-title":"Journal of Computational and Graphical Statistics"}],"container-title":["Journal of the Royal Statistical Society Series B: Statistical Methodology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/jrsssb\/article-pdf\/85\/4\/1326\/51801353\/qkad059.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/jrsssb\/article-pdf\/85\/4\/1326\/51801353\/qkad059.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,9,29]],"date-time":"2023-09-29T23:23:51Z","timestamp":1696029831000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/jrsssb\/article\/85\/4\/1326\/7193744"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,6,10]]},"references-count":40,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2023,6,10]]},"published-print":{"date-parts":[[2023,9,29]]}},"URL":"https:\/\/doi.org\/10.1093\/jrsssb\/qkad059","relation":{},"ISSN":["1369-7412","1467-9868"],"issn-type":[{"value":"1369-7412","type":"print"},{"value":"1467-9868","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2023,9]]},"published":{"date-parts":[[2023,6,10]]}}}