{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,24]],"date-time":"2026-09-24T04:29:50Z","timestamp":1790224190483,"version":"4.1.0"},"reference-count":55,"publisher":"Society for Industrial & Applied Mathematics (SIAM)","issue":"5","funder":[{"name":"Hong Kong Research Grant Council GRF","award":["12300519"],"award-info":[{"award-number":["12300519"]}]},{"name":"Hong Kong Research Grant Council GRF","award":["17201020"],"award-info":[{"award-number":["17201020"]}]},{"name":"Hong Kong Research Grant Council GRF","award":["17300021"],"award-info":[{"award-number":["17300021"]}]},{"name":"Hong Kong Research Grant Council CRF","award":["C1013-21GF"],"award-info":[{"award-number":["C1013-21GF"]}]},{"name":"Hong Kong Research Grant Council CRF","award":["C7004-21GF"],"award-info":[{"award-number":["C7004-21GF"]}]},{"name":"Joint NSFC-RGC","award":["N-HKU76921"],"award-info":[{"award-number":["N-HKU76921"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["SIAM J. Sci. Comput."],"published-print":{"date-parts":[[2023,10,31]]},"abstract":"<jats:p>Abstract.<\/jats:p>\n                  <jats:p>This paper studies deep neural networks for solving extremely large linear systems arising from high-dimensional problems. Because of the curse of dimensionality, it is expensive to store both the solution and right-hand side vector in such extremely large linear systems. Our idea is to employ a neural network to characterize the solution with many fewer parameters than the size of the solution under a matrix-free setting. We present an error analysis of the proposed method, indicating that the solution error is bounded by the condition number of the matrix and the neural network approximation error. Several numerical examples from partial differential equations, queueing problems, and probabilistic Boolean networks are presented to demonstrate that the solutions of linear systems can be learned quite accurately.<\/jats:p>","DOI":"10.1137\/22m1488132","type":"journal-article","created":{"date-parts":[[2023,9,21]],"date-time":"2023-09-21T04:50:07Z","timestamp":1695271807000},"page":"A2356-A2381","source":"Crossref","is-referenced-by-count":14,"title":["Deep Neural Networks for Solving Large Linear Systems Arising from High-Dimensional Problems"],"prefix":"10.1137","volume":"45","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-4341-8509","authenticated-orcid":true,"given":"Yiqi","family":"Gu","sequence":"first","affiliation":[{"name":"School of Mathematical Sciences, University of Electronic Science and Technology of China, Sichuan, 611731, China."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6833-5227","authenticated-orcid":true,"given":"Michael K.","family":"Ng","sequence":"additional","affiliation":[{"name":"Department of Mathematics, Hong Kong Baptist University, Kowloon Tong, Hong Kong."}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"351","published-online":{"date-parts":[[2023,9,21]]},"reference":[{"key":"ref1","unstructured":"Z. Allen-Zhu , \nY. Li , and \nY. Liang  , Learning and generalization in overparameterized neural networks, going beyond two layers, in Proceedings of the 33rd International Conference on Neural Information Processing Systems 2019, pp. 6158\u20136169."},{"key":"ref2","unstructured":"A. Andoni , \nR. Panigrahy , \nG. Valiant , and \nL. Zhang  , Learning polynomials with neural networks, in Proceedings of the 31st International Conference on Machine Learning, 2014."},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1016\/j.cam.2015.01.025"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1002\/nla.1818"},{"key":"ref5","unstructured":"A. R. Barron  , Neural net approximation, in Proceedings of the 7th Yale Workshop on Adaptive and Learning Systems, 1992."},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/18.256500"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1016\/S0024-3795(98)00015-9"},{"key":"ref8","doi-asserted-by":"crossref","unstructured":"Y. Cao , \nZ. Fang , \nY. Wu , \nD. X. Zhou , and \nQ. Gu  , Towards understanding the spectral bias of deep learning, in Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, 2021.","DOI":"10.24963\/ijcai.2021\/304"},{"key":"ref9","unstructured":"A. Caragea , \nP. Petersen , and \nF. Voigtlaender  , Neural Network Approximation and Estimation of Classifiers with Classification Boundary in a Barron Class, preprint, https:\/\/arxiv.org\/abs\/2011.09363, 2020."},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1007\/BF01396747"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1007\/BF01403891"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1137\/S0036144594276474"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1007\/BF02551274"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1007\/s10444-020-09790-2"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcp.2015.11.061"},{"key":"ref16","unstructured":"W. E, C. Ma , \nS. Wojtowytsch , and \nL. Wu  , Towards a Mathematical Understanding of Neural Network-Based Machine Learning: What We Know and What We Don\u2019t, preprint, https:\/\/arxiv.org\/abs\/2009.10713, 2020."},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.4310\/CMS.2019.v17.n5.a11"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1007\/s00365-021-09549-y"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1007\/s11425-018-9387-x"},{"key":"ref20","unstructured":"W. E  and \nS. Wojtowytsch  , Representation Formulas and Pointwise Properties for Barron Functions, preprint, https:\/\/arxiv.org\/abs\/2006.05982, 2020."},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1002\/nla.2106"},{"key":"ref22","unstructured":"X. Glorot  and \nY. Bengio  , Understanding the difficulty of training deep feedforward neural networks, in Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, 2010."},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.1718942115"},{"key":"ref24","doi-asserted-by":"crossref","unstructured":"K. He , \nX. Zhang , \nS. Ren , and \nJ. Sun  , Delving deep into rectifiers: Surpassing human-level performance on imagenet classification, in Proceedings of the IEEE International Conference on Computer Vision, 2015.","DOI":"10.1109\/ICCV.2015.123"},{"key":"ref25","doi-asserted-by":"crossref","unstructured":"K. He , \nX. Zhang , \nS. Ren , and \nJ. Sun  , Deep residual learning for image recognition, in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016.","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1016\/0893-6080(89)90020-8"},{"key":"ref27","unstructured":"X. Huang , \nX.L. Lin , \nM. K. Ng , and \nH.W. Sun  , Spectral Analysis for Preconditioning of Multi-dimensional Riesz Fractional Diffusion Equations, preprint, https:\/\/arxiv.org\/abs\/2102.01371, 2021."},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1137\/19M130981X"},{"key":"ref29","unstructured":"D. P. Kingma  and \nJ. Ba  , Adam: A Method for Stochastic Optimization, preprint, https:\/\/arxiv.org\/abs\/1412.6980, 2014."},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2018.2874447"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1137\/090756843"},{"key":"ref32","volume-title":"Partial Differential Equations with Numerical Methods","author":"Larsson S.","year":"2003"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1016\/j.cam.2012.02.022"},{"key":"ref34","unstructured":"S. Liang  and \nR. Srikant  , Why deep neural networks for function approximation? in Proceedings of the International Conference on Learning Representations, 2017."},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcp.2013.07.040"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1137\/20M134695X"},{"key":"ref37","unstructured":"Z. Lu , \nH. Pu , \nF. Wang , \nZ. Hu , and \nL. Wang  , The expressive power of neural networks: A view from the width, in Proceedings of the 31st International Conference on Neural Information Processing Systems, Curran Associates, Red Hook, NY, 2017, pp. 6232\u20136240."},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcp.2017.08.064"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1137\/18M1189336"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.4208\/jcm.2007-m2019-0239"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1016\/j.aml.2020.106294"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2018.08.019"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcp.2018.10.045"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2019.07.011"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.4208\/cicp.OA-2020-0149"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1162\/neco_a_01364"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2021.04.011"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2020.05.019"},{"key":"ref49","unstructured":"J. W. Siegel  and \nJ. Xu  , High-Order Approximation Rates for Neural Networks with   \\(\\text{ReLU}^k\\)   Activation Functions, preprint, https:\/\/arxiv.org\/abs\/2012.07205, 2020."},{"key":"ref50","unstructured":"T. Suzuki  , Adaptivity of deep ReLU network for learning in Besov and mixed smooth Besov spaces: Optimal rate and curse of dimensionality, in Proceedings of the International Conference on Learning Representations, 2018."},{"key":"ref51","unstructured":"R. Tipireddy , \nP. Perdikaris , \nP. Stinis , and \nA. Tartakovsky  , A Comparative Study of Physics-Informed Neural Network Models for Learning Unknown Dynamics and Constitutive Relations, preprint, https:\/\/arxiv.org\/abs\/1904.04058, 2019."},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2017.07.002"},{"key":"ref53","unstructured":"D. Yarotsky  , Optimal approximation of continuous functions by very deep ReLU networks, in Proceedings of the Conference on Learning Theory, 2018."},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcp.2020.109409"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1016\/j.aml.2018.12.022"}],"container-title":["SIAM Journal on Scientific Computing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/epubs.siam.org\/doi\/pdf\/10.1137\/22M1488132","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T15:46:13Z","timestamp":1787327173000},"score":1,"resource":{"primary":{"URL":"https:\/\/epubs.siam.org\/doi\/10.1137\/22M1488132"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,9,21]]},"references-count":55,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2023,10,31]]}},"alternative-id":["10.1137\/22M1488132"],"URL":"https:\/\/doi.org\/10.1137\/22m1488132","relation":{},"ISSN":["1064-8275","1095-7197"],"issn-type":[{"value":"1064-8275","type":"print"},{"value":"1095-7197","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,9,21]]}}}