{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,12]],"date-time":"2026-06-12T18:00:01Z","timestamp":1781287201261,"version":"3.54.1"},"reference-count":67,"publisher":"American Mathematical Society (AMS)","issue":"361","license":[{"start":{"date-parts":[[2026,7,18]],"date-time":"2026-07-18T00:00:00Z","timestamp":1784332800000},"content-version":"am","delay-in-days":365,"URL":"https:\/\/www.ams.org\/publications\/copyright-and-permissions"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Math. Comp."],"abstract":"

\n Inspired by the structure of spherical harmonics, we propose the truncated kernel stochastic gradient descent (T-kernel SGD) algorithm with a least-square loss function for spherical data fitting. T-kernel SGD introduces a novel regularization strategy by implementing SGD through a closed-form solution of the projection of the stochastic gradient in a low-dimensional subspace. In contrast to traditional kernel SGD, the regularization strategy implemented by T-kernel SGD is more effective in balancing bias and variance by dynamically adjusting the hypothesis space during iterations. The most significant advantage of the proposed algorithm is that it can achieve theoretically optimal convergence rates using a constant step size (independent of the sample size) while overcoming the inherent saturation problem of kernel SGD. Additionally, we leverage the structure of spherical polynomials to derive an equivalent T-kernel SGD, significantly reducing storage and computational costs compared to kernel SGD. Typically, T-kernel SGD requires only\n \n \n \n \n \n O<\/mml:mi>\n <\/mml:mrow>\n (<\/mml:mo>\n \n n<\/mml:mi>\n \n 1<\/mml:mn>\n +<\/mml:mo>\n \n d<\/mml:mi>\n \n d<\/mml:mi>\n \n \u2212\n \n <\/mml:mo>\n 1<\/mml:mn>\n <\/mml:mrow>\n <\/mml:mfrac>\n \n \u03f5\n \n <\/mml:mi>\n <\/mml:mrow>\n <\/mml:msup>\n )<\/mml:mo>\n <\/mml:mrow>\n \\mathcal {O}(n^{1+\\frac {d}{d-1}\\epsilon })<\/mml:annotation>\n <\/mml:semantics>\n <\/mml:math>\n <\/inline-formula>\n computational complexity and\n \n \n \n \n \n O<\/mml:mi>\n <\/mml:mrow>\n (<\/mml:mo>\n \n n<\/mml:mi>\n \n \n d<\/mml:mi>\n \n d<\/mml:mi>\n \n \u2212\n \n <\/mml:mo>\n 1<\/mml:mn>\n <\/mml:mrow>\n <\/mml:mfrac>\n \n \u03f5\n \n <\/mml:mi>\n <\/mml:mrow>\n <\/mml:msup>\n )<\/mml:mo>\n <\/mml:mrow>\n \\mathcal {O}(n^{\\frac {d}{d-1}\\epsilon })<\/mml:annotation>\n <\/mml:semantics>\n <\/mml:math>\n <\/inline-formula>\n storage to achieve optimal rates for the d-dimensional sphere, where\n \n \n \n \n 0<\/mml:mn>\n ><\/mml:mo>\n \n \u03f5\n \n <\/mml:mi>\n ><\/mml:mo>\n \n 1<\/mml:mn>\n 2<\/mml:mn>\n <\/mml:mfrac>\n <\/mml:mrow>\n 0>\\epsilon >\\frac {1}{2}<\/mml:annotation>\n <\/mml:semantics>\n <\/mml:math>\n <\/inline-formula>\n can be arbitrarily small if the optimal fitting or the underlying space possesses sufficient regularity. This regularity is determined by the smoothness parameter of the objective function and the decaying rate of the eigenvalues of the integral operator associated with the kernel function, both of which reflect the difficulty of the estimation problem. Our main results quantitatively characterize how this prior information influences the convergence of T-kernel SGD. The numerical experiments further validate the theoretical findings presented in this paper.\n <\/p>","DOI":"10.1090\/mcom\/4124","type":"journal-article","created":{"date-parts":[[2025,7,18]],"date-time":"2025-07-18T18:24:17Z","timestamp":1752863057000},"page":"2409-2480","source":"Crossref","is-referenced-by-count":0,"title":["Truncated kernel stochastic gradient descent on spheres"],"prefix":"10.1090","volume":"95","author":[{"given":"Jinhui","family":"Bai","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lei","family":"Shi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"14","published-online":{"date-parts":[[2025,7,18]]},"reference":[{"key":"1","unstructured":"A. Abedsoltan, M. Belkin, and P. Pandit, Toward large kernel models, Proceedings of the 40th International Conference on Machine Learning, Proceedings of Machine Learning Research, vol. 202, pp. 61\u201378, PMLR, 2023."},{"key":"2","unstructured":"A. Abedsoltan, S. Ma, P. Pandit, and M. Belkin. Fast training of large kernel models with delayed projections, Preprint, arXiv:2411.16658, 2024."},{"issue":"1","key":"3","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1137\/21M1425050","article-title":"Kernel approximation on algebraic varieties","volume":"7","author":"Altschuler, Jason M.","year":"2023","journal-title":"SIAM J. Appl. Algebra Geom."},{"issue":"3","key":"4","doi-asserted-by":"publisher","first-page":"177","DOI":"10.1016\/S1063-5203(02)00507-9","article-title":"Wavelets on the sphere: implementation and approximations","volume":"13","author":"Antoine, J.-P.","year":"2002","journal-title":"Appl. Comput. Harmon. Anal.","ISSN":"https:\/\/id.crossref.org\/issn\/1063-5203","issn-type":"print"},{"issue":"4","key":"5","doi-asserted-by":"publisher","first-page":"1116","DOI":"10.1137\/16M1105396","article-title":"Faster kernel ridge regression using sketching and preconditioning","volume":"38","author":"Avron, Haim","year":"2017","journal-title":"SIAM J. Matrix Anal. Appl.","ISSN":"https:\/\/id.crossref.org\/issn\/0895-4798","issn-type":"print"},{"key":"6","unstructured":"F. Bach, Sharp analysis of low-rank kernel matrix approximations, Conference on Learning Theory, pp. 185\u2013209, PMLR, 2013."},{"key":"7","unstructured":"R. Berthier, F. Bach, and P. Gaillard, Tight nonparametric convergence rates for stochastic gradient descent under the noiseless linear model, Advances in Neural Information Processing Systems, vol. 33, pp. 2576\u20132586, Curran Associates, Inc., 2020."},{"key":"8","unstructured":"G. Blanchard and N. Kr\u00e4mer, Optimal learning rates for kernel conjugate gradient regression. Advances in Neural Information Processing Systems, vol. 23, Curran Associates, Inc., 2010."},{"issue":"3","key":"9","doi-asserted-by":"publisher","first-page":"331","DOI":"10.1007\/s10208-006-0196-8","article-title":"Optimal rates for the regularized least-squares algorithm","volume":"7","author":"Caponnetto, A.","year":"2007","journal-title":"Found. Comput. Math.","ISSN":"https:\/\/id.crossref.org\/issn\/1615-3375","issn-type":"print"},{"key":"10","isbn-type":"print","doi-asserted-by":"crossref","DOI":"10.1137\/1.9781611972597","volume-title":"Linear and nonlinear functional analysis with applications","author":"Ciarlet, Philippe G.","year":"2013","ISBN":"https:\/\/id.crossref.org\/isbn\/9781611972580"},{"key":"11","doi-asserted-by":"crossref","unstructured":"F. Cucker and S. Smale, On the mathematical foundations of learning, Bulletin of the American Mathematical Society. New Series, 39(1):1\u201349, 2002.","DOI":"10.1090\/S0273-0979-01-00923-5"},{"key":"12","series-title":"Springer Monographs in Mathematics","isbn-type":"print","doi-asserted-by":"publisher","DOI":"10.1007\/978-1-4614-6660-4","volume-title":"Approximation theory and harmonic analysis on spheres and balls","author":"Dai, Feng","year":"2013","ISBN":"https:\/\/id.crossref.org\/isbn\/9781461466598"},{"key":"13","doi-asserted-by":"crossref","unstructured":"A. Defazio, X. Yang, A. Khaled, K. Mishchenko, H. Mehta, and A. Cutkosky, The road less scheduled. Advances in Neural Information Processing Systems, vol. 37, pp. 9974\u201310007, Curran Associates, Inc., 2024.","DOI":"10.52202\/079017-0320"},{"key":"14","doi-asserted-by":"publisher","first-page":"30","DOI":"10.1016\/j.spl.2018.07.019","article-title":"Local binary regression with spherical predictors","volume":"144","author":"Di Marzio, Marco","year":"2019","journal-title":"Statist. Probab. Lett.","ISSN":"https:\/\/id.crossref.org\/issn\/0167-7152","issn-type":"print"},{"issue":"506","key":"15","doi-asserted-by":"publisher","first-page":"748","DOI":"10.1080\/01621459.2013.866567","article-title":"Nonparametric regression for spherical data","volume":"109","author":"Di Marzio, Marco","year":"2014","journal-title":"J. Amer. Statist. Assoc.","ISSN":"https:\/\/id.crossref.org\/issn\/0162-1459","issn-type":"print"},{"key":"16","unstructured":"M. D\u00edaz, E. N. Epperly, Z. Frangella, J. A. Tropp, and R. J. Webber, Robust, randomized preconditioning for kernel ridge regression, Preprint, arXiv:2304.12465, 2023."},{"issue":"4","key":"17","doi-asserted-by":"publisher","first-page":"1363","DOI":"10.1214\/15-AOS1391","article-title":"Nonparametric stochastic approximation with large step-sizes","volume":"44","author":"Dieuleveut, Aymeric","year":"2016","journal-title":"Ann. Statist.","ISSN":"https:\/\/id.crossref.org\/issn\/0090-5364","issn-type":"print"},{"issue":"3","key":"18","doi-asserted-by":"publisher","first-page":"1348","DOI":"10.1214\/19-AOS1850","article-title":"Bridging the gap between constant step size stochastic gradient descent and Markov chains","volume":"48","author":"Dieuleveut, Aymeric","year":"2020","journal-title":"Ann. Statist.","ISSN":"https:\/\/id.crossref.org\/issn\/0090-5364","issn-type":"print"},{"key":"19","isbn-type":"print","first-page":"117","article-title":"Scattered data fitting on the sphere","author":"Fasshauer, Gregory E.","year":"1998","ISBN":"https:\/\/id.crossref.org\/isbn\/0826513158"},{"key":"20","doi-asserted-by":"publisher","first-page":"295","DOI":"10.1098\/rspa.1953.0064","article-title":"Dispersion on a sphere","volume":"217","author":"Fisher, Ronald","year":"1953","journal-title":"Proc. Roy. Soc. London Ser. A","ISSN":"https:\/\/id.crossref.org\/issn\/0962-8444","issn-type":"print"},{"key":"21","series-title":"Numerical Mathematics and Scientific Computation","isbn-type":"print","doi-asserted-by":"crossref","DOI":"10.1093\/oso\/9780198536826.001.0001","volume-title":"Constructive approximation on the sphere","author":"Freeden, W.","year":"1998","ISBN":"https:\/\/id.crossref.org\/isbn\/0198536828"},{"issue":"1","key":"22","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1214\/aos\/1176347963","article-title":"Multivariate adaptive regression splines","volume":"19","author":"Friedman, Jerome H.","year":"1991","journal-title":"Ann. Statist.","ISSN":"https:\/\/id.crossref.org\/issn\/0090-5364","issn-type":"print"},{"issue":"4","key":"23","doi-asserted-by":"publisher","first-page":"1178","DOI":"10.1111\/sjos.12236","article-title":"Testing parametric models in linear-directional regression","volume":"43","author":"Garc\u00eda-Portugu\u00e9s, Eduardo","year":"2016","journal-title":"Scand. J. Stat.","ISSN":"https:\/\/id.crossref.org\/issn\/0303-6898","issn-type":"print"},{"issue":"2","key":"24","doi-asserted-by":"publisher","first-page":"227","DOI":"10.1080\/10618600.2019.1652616","article-title":"Scalable Gaussian process computations using hierarchical matrices","volume":"29","author":"Geoga, Christopher J.","year":"2020","journal-title":"J. Comput. Graph. Statist.","ISSN":"https:\/\/id.crossref.org\/issn\/1061-8600","issn-type":"print"},{"issue":"5-6","key":"25","doi-asserted-by":"publisher","first-page":"2745","DOI":"10.1007\/s10444-019-09707-8","article-title":"Fast and strong convergence of online learning algorithms","volume":"45","author":"Guo, Zheng-Chu","year":"2019","journal-title":"Adv. Comput. Math.","ISSN":"https:\/\/id.crossref.org\/issn\/1019-7168","issn-type":"print"},{"issue":"7","key":"26","doi-asserted-by":"publisher","first-page":"1314","DOI":"10.1002\/cpa.21577","article-title":"Hierarchical interpolative factorization for elliptic operators: integral equations","volume":"69","author":"Ho, Kenneth L.","year":"2016","journal-title":"Comm. Pure Appl. Math.","ISSN":"https:\/\/id.crossref.org\/issn\/0010-3640","issn-type":"print"},{"key":"27","series-title":"SpringerBriefs in Mathematics","isbn-type":"print","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-319-17939-1","volume-title":"Spherical radial basis functions, theory and applications","author":"Hubbert, Simon","year":"2015","ISBN":"https:\/\/id.crossref.org\/isbn\/9783319179384"},{"key":"28","series-title":"Universitext","isbn-type":"print","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-61860-9","volume-title":"Riemannian geometry and geometric analysis","author":"Jost, J\u00fcrgen","year":"2017","ISBN":"https:\/\/id.crossref.org\/isbn\/9783319618593","edition":"7"},{"issue":"1","key":"29","doi-asserted-by":"publisher","first-page":"124","DOI":"10.1214\/19-STS755","article-title":"A general framework for Vecchia approximations of Gaussian processes","volume":"36","author":"Katzfuss, Matthias","year":"2021","journal-title":"Statist. Sci.","ISSN":"https:\/\/id.crossref.org\/issn\/0883-4237","issn-type":"print"},{"issue":"3","key":"30","doi-asserted-by":"publisher","first-page":"383","DOI":"10.1007\/s13253-020-00401-7","article-title":"Vecchia approximations of Gaussian-process predictions","volume":"25","author":"Katzfuss, Matthias","year":"2020","journal-title":"J. Agric. Biol. Environ. Stat.","ISSN":"https:\/\/id.crossref.org\/issn\/1085-7117","issn-type":"print"},{"issue":"8","key":"31","doi-asserted-by":"publisher","first-page":"2165","DOI":"10.1109\/TSP.2004.830991","article-title":"Online learning with kernels","volume":"52","author":"Kivinen, Jyrki","year":"2004","journal-title":"IEEE Trans. Signal Process.","ISSN":"https:\/\/id.crossref.org\/issn\/1053-587X","issn-type":"print"},{"key":"32","first-page":"981","article-title":"Sampling methods for the Nystr\u00f6m method","volume":"13","author":"Kumar, Sanjiv","year":"2012","journal-title":"J. Mach. Learn. Res.","ISSN":"https:\/\/id.crossref.org\/issn\/1532-4435","issn-type":"print"},{"key":"33","doi-asserted-by":"publisher","first-page":"66","DOI":"10.1016\/j.jco.2018.09.003","article-title":"Nonparametric regression using needlet kernels for spherical data","volume":"50","author":"Lin, Shao-Bo","year":"2019","journal-title":"J. Complexity","ISSN":"https:\/\/id.crossref.org\/issn\/0885-064X","issn-type":"print"},{"issue":"3","key":"34","doi-asserted-by":"publisher","first-page":"1098","DOI":"10.1137\/22M1519948","article-title":"Kernel interpolation of high dimensional scattered data","volume":"62","author":"Lin, Shao-Bo","year":"2024","journal-title":"SIAM J. Numer. Anal.","ISSN":"https:\/\/id.crossref.org\/issn\/0036-1429","issn-type":"print"},{"key":"35","first-page":"Paper No. 92, 31","article-title":"Distributed learning with regularized least squares","volume":"18","author":"Lin, Shao-Bo","year":"2017","journal-title":"J. Mach. Learn. Res.","ISSN":"https:\/\/id.crossref.org\/issn\/1532-4435","issn-type":"print"},{"issue":"1","key":"36","doi-asserted-by":"publisher","first-page":"A313--A337","DOI":"10.1137\/22M1484377","article-title":"Sketching with spherical designs for noisy data fitting on spheres","volume":"46","author":"Lin, Shao-Bo","year":"2024","journal-title":"SIAM J. Sci. Comput.","ISSN":"https:\/\/id.crossref.org\/issn\/1064-8275","issn-type":"print"},{"issue":"4","key":"37","doi-asserted-by":"publisher","first-page":"423","DOI":"10.1111\/j.1467-9868.2011.00777.x","article-title":"An explicit link between Gaussian fields and Gaussian Markov random fields: the stochastic partial differential equation approach","volume":"73","author":"Lindgren, Finn","year":"2011","journal-title":"J. R. Stat. Soc. Ser. B Stat. Methodol.","ISSN":"https:\/\/id.crossref.org\/issn\/1369-7412","issn-type":"print"},{"key":"38","doi-asserted-by":"publisher","first-page":"115","DOI":"10.1016\/j.csda.2019.02.002","article-title":"Likelihood approximation with hierarchical matrices for large spatial datasets","volume":"137","author":"Litvinenko, Alexander","year":"2019","journal-title":"Comput. Statist. Data Anal.","ISSN":"https:\/\/id.crossref.org\/issn\/0167-9473","issn-type":"print"},{"key":"39","doi-asserted-by":"crossref","unstructured":"F. Liu, X. Huang, Y. Chen, and J. A. K. Suykens, Random features for kernel approximation: a survey on algorithms, theory, and beyond, IEEE Trans. Pattern Anal. Mach. Intell. 44 (2022), no. 10, 7128\u20137148.","DOI":"10.1109\/TPAMI.2021.3097011"},{"issue":"7","key":"40","doi-asserted-by":"publisher","first-page":"075002","DOI":"10.1088\/1361-6420\/ab129e","article-title":"Nystr\u00f6m subsampling method for coefficient-based regularized regression","volume":"35","author":"Ma, Longda","year":"2019","journal-title":"Inverse Problems","ISSN":"https:\/\/id.crossref.org\/issn\/0266-5611","issn-type":"print"},{"key":"41","series-title":"Springer Series in Statistics","isbn-type":"print","doi-asserted-by":"publisher","DOI":"10.1007\/978-0-387-68276-1","volume-title":"Inequalities: theory of majorization and its applications","author":"Marshall, Albert W.","year":"2011","ISBN":"https:\/\/id.crossref.org\/isbn\/9780387400877","edition":"2"},{"key":"42","unstructured":"U. Marteau-Ferey, D. Ostrovskii, F. Bach, and A. Rudi, Beyond least-squares: fast rates for regularized empirical risk minimization through self-concordance, Proceedings of the Thirty-Second Conference on Learning Theory, Proceedings of Machine Learning Research, vol. 99, pp. 2294\u20132340, PMLR, 2019."},{"key":"43","unstructured":"N. Muecke, G. Neu, and L. Rosasco, Beating SGD saturation with tail-averaging and minibatching, Advances in Neural Information Processing Systems, vol. 32, Curran Associates, Inc., 2019."},{"key":"44","unstructured":"C. Musco and C. Musco, Recursive sampling for the Nystr\u00f6m method, Advances in Neural Information Processing Systems, vol. 30, Curran Associates, Inc., 2017."},{"issue":"6","key":"45","doi-asserted-by":"publisher","first-page":"1393","DOI":"10.1137\/S0036141001395054","article-title":"Scattered data interpolation on spheres: error estimates and locally supported basis functions","volume":"33","author":"Narcowich, Francis J.","year":"2002","journal-title":"SIAM J. Math. Anal.","ISSN":"https:\/\/id.crossref.org\/issn\/0036-1410","issn-type":"print"},{"key":"46","unstructured":"A. Rahimi and B. Recht, Random features for large-scale kernel machines, Advances in Neural Information Processing Systems, vol. 20, Curran Associates, Inc., 2007."},{"key":"47","unstructured":"P. Rathore, Z. Frangella, and M. Udell, Have ASkotch: fast methods for large-scale, memory-constrained kernel ridge regression, Preprint arXiv:2407.10070, 2024."},{"key":"48","doi-asserted-by":"publisher","first-page":"400","DOI":"10.1214\/aoms\/1177729586","article-title":"A stochastic approximation method","volume":"22","author":"Robbins, Herbert","year":"1951","journal-title":"Ann. Math. Statistics","ISSN":"https:\/\/id.crossref.org\/issn\/0003-4851","issn-type":"print"},{"key":"49","unstructured":"L. Rosasco and S. Villa, Learning with incremental iterative regularization, Advances in Neural Information Processing Systems, vol. 28, Curran Associates, Inc., 2015."},{"key":"50","unstructured":"A. Rudi, D. Calandriello, L. Carratino, and L. Rosasco, On fast leverage score sampling and optimal learning. Advances in Neural Information Processing Systems, vol. 31, Curran Associates, Inc., 2018."},{"key":"51","unstructured":"A. Rudi, L. Carratino, and L. Rosasco, FALKON: an optimal large scale kernel method. Advances in Neural Information Processing Systems, vol. 30, Curran Associates, Inc., 2017."},{"issue":"3","key":"52","doi-asserted-by":"publisher","first-page":"A2019--A2046","DOI":"10.1137\/20M1336254","article-title":"Sparse Cholesky factorization by Kullback-Leibler minimization","volume":"43","author":"Sch\u00e4fer, Florian","year":"2021","journal-title":"SIAM J. Sci. Comput.","ISSN":"https:\/\/id.crossref.org\/issn\/1064-8275","issn-type":"print"},{"key":"53","unstructured":"L. Shi and J.-Q. Yang, Learning operators with stochastic gradient descent in general hilbert spaces, Preprint, arXiv:2402.04691, 2024."},{"issue":"6","key":"54","doi-asserted-by":"publisher","first-page":"1095","DOI":"10.1142\/S0219530524500131","article-title":"Iterative kernel regression with preconditioning","volume":"22","author":"Shi, Lei","year":"2024","journal-title":"Anal. Appl. (Singap.)","ISSN":"https:\/\/id.crossref.org\/issn\/0219-5305","issn-type":"print"},{"issue":"2","key":"55","doi-asserted-by":"publisher","first-page":"145","DOI":"10.1007\/s10208-004-0160-z","article-title":"Online learning algorithms","volume":"6","author":"Smale, Steve","year":"2006","journal-title":"Found. Comput. Math.","ISSN":"https:\/\/id.crossref.org\/issn\/1615-3375","issn-type":"print"},{"issue":"1","key":"56","doi-asserted-by":"publisher","first-page":"17","DOI":"10.1142\/S0219530503000089","article-title":"Estimating the approximation error in learning theory","volume":"1","author":"Smale, Steve","year":"2003","journal-title":"Anal. Appl. (Singap.)","ISSN":"https:\/\/id.crossref.org\/issn\/0219-5305","issn-type":"print"},{"key":"57","unstructured":"A. J. Smola and B. Sch\u00f6lkopf, Sparse greedy matrix approximation for machine learning, Proceedings of the Seventeenth International Conference on Machine Learning (ICML 2000) (Stanford University, Stanford, CA, USA, June 29\u2013July 2, 2000), pp. 911\u2013918, Morgan Kaufmann, 2000."},{"issue":"9","key":"58","doi-asserted-by":"publisher","first-page":"5716","DOI":"10.1109\/TIT.2014.2332531","article-title":"Online learning as stochastic approximation of regularization paths: optimality and almost-sure convergence","volume":"60","author":"Tarr\u00e8s, Pierre","year":"2014","journal-title":"IEEE Trans. Inform. Theory","ISSN":"https:\/\/id.crossref.org\/issn\/0018-9448","issn-type":"print"},{"issue":"1","key":"59","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1137\/0902002","article-title":"Spline interpolation and smoothing on the sphere","volume":"2","author":"Wahba, Grace","year":"1981","journal-title":"SIAM J. Sci. Statist. Comput.","ISSN":"https:\/\/id.crossref.org\/issn\/0196-5204","issn-type":"print"},{"key":"60","series-title":"Cambridge Series in Statistical and Probabilistic Mathematics","isbn-type":"print","doi-asserted-by":"publisher","DOI":"10.1017\/9781108627771","volume-title":"High-dimensional statistics","volume":"48","author":"Wainwright, Martin J.","year":"2019","ISBN":"https:\/\/id.crossref.org\/isbn\/9781108498029"},{"key":"61","series-title":"Cambridge Monographs on Applied and Computational Mathematics","isbn-type":"print","volume-title":"Scattered data approximation","volume":"17","author":"Wendland, Holger","year":"2005","ISBN":"https:\/\/id.crossref.org\/isbn\/9780521843355"},{"key":"62","unstructured":"C. Williams and M. Seeger, Using the Nystr\u00f6m method to speed up kernel machines, Advances in Neural Information Processing Systems, vol. 13, MIT Press, 2000."},{"issue":"2","key":"63","doi-asserted-by":"publisher","first-page":"289","DOI":"10.1007\/s00365-006-0663-2","article-title":"On early stopping in gradient descent learning","volume":"26","author":"Yao, Yuan","year":"2007","journal-title":"Constr. Approx.","ISSN":"https:\/\/id.crossref.org\/issn\/0176-4276","issn-type":"print"},{"issue":"5","key":"64","doi-asserted-by":"publisher","first-page":"561","DOI":"10.1007\/s10208-006-0237-y","article-title":"Online gradient descent learning algorithms","volume":"8","author":"Ying, Yiming","year":"2008","journal-title":"Found. Comput. Math.","ISSN":"https:\/\/id.crossref.org\/issn\/1615-3375","issn-type":"print"},{"issue":"5","key":"65","doi-asserted-by":"publisher","first-page":"2848","DOI":"10.1214\/22-aos2212","article-title":"A sieve stochastic gradient descent estimator for online nonparametric regression in Sobolev ellipsoids","volume":"50","author":"Zhang, Tianyu","year":"2022","journal-title":"Ann. Statist.","ISSN":"https:\/\/id.crossref.org\/issn\/0090-5364","issn-type":"print"},{"key":"66","first-page":"3299","article-title":"Divide and conquer kernel ridge regression: a distributed algorithm with minimax optimal rates","volume":"16","author":"Zhang, Yuchen","year":"2015","journal-title":"J. Mach. Learn. Res.","ISSN":"https:\/\/id.crossref.org\/issn\/1532-4435","issn-type":"print"},{"key":"67","doi-asserted-by":"publisher","first-page":"Paper No. 107081, 17","DOI":"10.1016\/j.csda.2020.107081","article-title":"Vecchia-Laplace approximations of generalized Gaussian processes for big non-Gaussian spatial data","volume":"153","author":"Zilber, Daniel","year":"2021","journal-title":"Comput. Statist. Data Anal.","ISSN":"https:\/\/id.crossref.org\/issn\/0167-9473","issn-type":"print"}],"container-title":["Mathematics of Computation"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.ams.org\/mcom\/2026-95-361\/S0025-5718-2025-04124-3\/mcom4124_AM.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/www.ams.org\/mcom\/2026-95-361\/S0025-5718-2025-04124-3\/S0025-5718-2025-04124-3.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,12]],"date-time":"2026-06-12T17:54:01Z","timestamp":1781286841000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.ams.org\/mcom\/2026-95-361\/S0025-5718-2025-04124-3\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7,18]]},"references-count":67,"journal-issue":{"issue":"361","published-print":{"date-parts":[[2026,9]]}},"alternative-id":["S0025-5718-2025-04124-3"],"URL":"https:\/\/doi.org\/10.1090\/mcom\/4124","archive":["CLOCKSS","Portico"],"relation":{},"ISSN":["1088-6842","0025-5718"],"issn-type":[{"value":"1088-6842","type":"electronic"},{"value":"0025-5718","type":"print"}],"subject":[],"published":{"date-parts":[[2025,7,18]]}}}