{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,18]],"date-time":"2026-09-18T08:37:59Z","timestamp":1789720679236,"version":"4.0.1"},"reference-count":55,"publisher":"Society for Industrial & Applied Mathematics (SIAM)","issue":"4","funder":[{"DOI":"10.13039\/501100012166","name":"National Key Research and Development Program of China","doi-asserted-by":"publisher","award":["2021YFA1001300"],"award-info":[{"award-number":["2021YFA1001300"]}],"id":[{"id":"10.13039\/501100012166","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["11901338"],"award-info":[{"award-number":["11901338"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Tsinghua University Initiative Scientific Research Program"},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["12071244"],"award-info":[{"award-number":["12071244"]}],"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":["12090022"],"award-info":[{"award-number":["12090022"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Beijing Natural Science Foundation","award":["12090022"],"award-info":[{"award-number":["12090022"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["SIAM J. Sci. Comput."],"published-print":{"date-parts":[[2023,8,31]]},"abstract":"<jats:p>Abstract.<\/jats:p>\n                  <jats:p>The dynamic formulation of optimal transport has attracted growing interest in scientific computing and machine learning, and its computation requires solving a PDE-constrained optimization problem. The classical Eulerian discretization based approaches suffer from the curse of dimensionality, which arises from the approximation of a high-dimensional velocity field. In this work, we propose a deep learning based method for solving the dynamic optimal transport in high-dimensional space. Our method contains three main ingredients: a carefully designed representation of the velocity field, the discretization of the PDE constraint along the characteristics, and the computation of a high-dimensional integral by the Monte Carlo method in each time step. Specifically, in the representation of the velocity field, we apply the classical nodal basis function in time and the deep neural networks in the space domain with the H1-norm regularization. This technique promotes the regularity of the velocity field in both time and space such that the discretization along the characteristic remains stable during the training process. Extensive numerical examples have been conducted to test the proposed method. Compared to other solvers of optimal transport, our method could give more accurate results in high-dimensional cases and has very good scalability with respect to dimension. Finally, we extend our method to more complicated cases such as the crowd motion problem.<\/jats:p>","DOI":"10.1137\/22m1490831","type":"journal-article","created":{"date-parts":[[2023,8,1]],"date-time":"2023-08-01T04:30:36Z","timestamp":1690864236000},"page":"B544-B563","source":"Crossref","is-referenced-by-count":6,"title":["A Scalable Deep Learning Approach for Solving High-Dimensional Dynamic Optimal Transport"],"prefix":"10.1137","volume":"45","author":[{"given":"Wei","family":"Wan","sequence":"first","affiliation":[{"name":"School of Mathematics and Physics, North China Electric Power University, Beijing, China."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuejin","family":"Zhang","sequence":"additional","affiliation":[{"name":"Department of Mathematical Sciences, Tsinghua University, Beijing 100084, China."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chenglong","family":"Bao","sequence":"additional","affiliation":[{"name":"Yau Mathematical Sciences Center, Tsinghua University, Beijing 100084, China, and Yanqi Lake Beijing Institute of Mathematical Sciences and Applications, Beijing 101408, China."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bin","family":"Dong","sequence":"additional","affiliation":[{"name":"Beijing International Center for Mathematical Research and Center for Machine Learning Research, Peking University, Beijing, China."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9122-0302","authenticated-orcid":true,"given":"Zuoqiang","family":"Shi","sequence":"additional","affiliation":[{"name":"Yau Mathematical Sciences Center, Tsinghua University, Beijing 100084, China, and Yanqi Lake Beijing Institute of Mathematical Sciences and Applications, Beijing 101408, China."}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"351","published-online":{"date-parts":[[2023,8,1]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2014.2339736"},{"key":"ref2","unstructured":"B. Amos , \nL. Xu , and \nJ. Z. Kolter , Input convex neural networks, in Proceedings of the 34th International Conference on Machine Learning, PMLR, 2017, pp. 146\u2013155."},{"key":"ref3","unstructured":"D. An , \nY. Guo , \nN. Lei , \nZ. Luo , \nS.T. Yau , and \nX. Gu , AE-OT: A new generative model based on extended semi-discrete optimal transport, in Proceedings of the International Conference on Learning Representations 2020."},{"key":"ref4","doi-asserted-by":"crossref","unstructured":"D. An , \nY. Guo , \nM. Zhang , \nX. Qi , \nN. Lei , and \nX. Gu , AE-OT-GAN: Training GANs from data specific latent distribution, in Proceedings of the 16th European Conference on Computer Vision, Springer, 2020, pp. 548\u2013564.","DOI":"10.1007\/978-3-030-58574-7_33"},{"key":"ref5","unstructured":"M. Arjovsky , \nS. Chintala , and \nL. Bottou , Wasserstein GAN, preprint, https:\/\/arxiv.org\/abs\/1701.07875, 2017."},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1007\/s002110050002"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcp.2013.12.015"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1002\/cpa.3160440402"},{"key":"ref9","doi-asserted-by":"crossref","unstructured":"R. Collobert  and \nJ. Weston , A unified architecture for natural language processing: Deep neural networks with multitask learning, in Proceedings of the 25th international Conference on Machine Learning, PMLR, 2008, pp. 160\u2013167.","DOI":"10.1145\/1390156.1390177"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2016.2615921"},{"key":"ref11","unstructured":"M. Cuturi , Sinkhorn distances: Lightspeed computation of optimal transport, in NIPS\u201913: Proceedings of the 26th International Conference on Neural Information Processing Systems - Volume 2, Curran Associates, 2013, pp. 2292\u20132300."},{"key":"ref12","doi-asserted-by":"crossref","unstructured":"L. Deng , \nG. Hinton , and \nB. Kingsbury , New types of deep neural network learning for speech recognition and related applications: An overview, in Proceedings of the 2013 IEEE International Conference on Acoustics, Speech and Signal Processing, IEEE, 2013, pp. 8599\u20138603.","DOI":"10.1109\/ICASSP.2013.6639344"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1017\/S0962492921000052"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1007\/s40304-017-0117-6"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1007\/s40304-018-0127-z"},{"key":"ref16","unstructured":"C. Finlay , \nJ.H. Jacobsen , \nL. Nurbekyan , and \nA. Oberman , How to train your neural ODE: The world of Jacobian and kinetic regularization, in Proceedings of the 37th International Conference on Machine Learning, PMLR, 2020, pp. 3154\u20133164."},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcp.2019.108940"},{"key":"ref18","unstructured":"A. Genevay , \nM. Cuturi , \nG. Peyr\u00e9 , and \nF. Bach , Stochastic optimization for large-scale optimal transport, in NIPS \u201916: Proceedings of the 30th International Conference on Neural Information Processing Systems, Curran Associates, 2016, pp. 3440\u20133448."},{"key":"ref19","doi-asserted-by":"crossref","unstructured":"A. Gholami , \nK. Keutzer , and \nG. Biros , ANODE: Unconditionally accurate memory-efficient gradients for neural ODEs, in Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence (IJCAI \u201919), International Joint Conferences on Artificial Intelligence, 2019, pp. 730\u2013736.","DOI":"10.24963\/ijcai.2019\/103"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1613\/jair.4992"},{"key":"ref21","volume-title":"Deep Learning","author":"Goodfellow I.","year":"2016"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.4310\/AJM.2016.v20.n2.a7"},{"key":"ref23","unstructured":"M. Henry , \nE. Maitre , and \nV. Perrier , Primal-Dual Formulation of the Dynamic Optimal Transport Using Helmholtz-Hodge Decomposition, preprint, hal-02087017f, 2019."},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1287\/mnsc.5.1.1"},{"key":"ref25","unstructured":"A. Krizhevsky , \nI. Sutskever , and \nG. E. Hinton , ImageNet classification with deep convolutional neural networks, in Advances in Neural Information Processing Systems 25 (NIPS 2012), Curran Associates, 2012, pp. 1097\u20131105."},{"key":"ref26","unstructured":"M. Kusner , \nY. Sun , \nN. Kolkin , and \nK. Weinberger , From word embeddings to document distances, in ICML\u201915: Proceedings of the 32nd International Conference on Machine Learning, Vol. 37, JMLR.org, 2015, pp. 957\u2013966."},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1007\/s11537-007-0657-8"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1016\/j.cagd.2018.10.005"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1137\/16M1071560"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1137\/18M1219813"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1137\/20M134695X"},{"key":"ref32","unstructured":"A. V. Makkuva , \nA. Taghvaei , \nS. Oh , and \nJ. Lee , Optimal transport mapping via input convex neural networks, in ICML\u201920: Proceedings of the   \\(37\\)  th International Conference on Machine Learning, JMLR.org, 2020, pp. 6672\u20136681."},{"key":"ref33","unstructured":"G. Monge , M\u00e9moire sur la th\u00e9orie des d\u00e9blais et des remblais, Histoire de l\u2019Acad\u00e9mie Royale des Sciences, 1781."},{"key":"ref34","unstructured":"D. Onken  and \nL. Ruthotto , Discretize-Optimize vs. Optimize-Discretize for Time-Series Regression and Continuous Normalizing Flows, preprint, https:\/\/arxiv.org\/abs\/2005.13420v1, 2020."},{"key":"ref35","unstructured":"N. Papadakis , Optimal Transport for Image Processing, Habilitation thesis, Universit\u00e9 de Bordeaux, 2015."},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1137\/130920058"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1561\/2200000073"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1137\/140986414"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcp.2018.10.045"},{"key":"ref40","unstructured":"I. Redko , \nN. Courty , \nR. Flamary , and \nD. Tuia , Optimal Transport for Multi-source Domain Adaptation Under Target Shift, preprint, arXiv:1803.04899, 2019."},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.1922204117"},{"key":"ref42","unstructured":"T. Salimans , \nH. Zhang , \nA. Radford , and \nD. Metaxas , Improving GANs Using Optimal Transport, preprint, arXiv:1803.05573, 2018."},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-20828-2"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2014.09.003"},{"key":"ref45","unstructured":"V. Seguy , \nB. B. Damodaran , \nR. Flamary , \nN. Courty , \nA. Rolet , and \nM. Blondel , Large Scale Optimal Transport and Mapping Estimation, preprint, arXiv:1711.02283, 2017."},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2021.04.011"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1214\/aoms\/1177703591"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcp.2018.08.029"},{"key":"ref49","doi-asserted-by":"crossref","unstructured":"F. Sultana , \nA. Sufian , and \nP. Dutta , Advancements in image classification using convolutional neural network, in Proceedings of the 2018 Fourth International Conference on Research in Computational Intelligence and Communication Networks (ICRCICN), IEEE, 2018, pp. 122\u2013129.","DOI":"10.1109\/ICRCICN.2018.8718718"},{"key":"ref50","unstructured":"A. Tong , \nJ. Huang , \nG. Wolf , \nD. V. Dijk , and \nS. Krishnaswamy , Trajectorynet: A dynamic optimal transport network for modeling cellular dynamics, in Proceedings of the 37th International Conference on Machine Learning, PMLR 119, JMLR.org, 2020, pp. 9526\u20139536."},{"key":"ref51","unstructured":"L. C. Torres , \nL. M. Pereira , and \nM. H. Amini , A Survey on Optimal Transport for Machine Learning: Theory and Applications, preprint, arXiv:2106.01963, 2021."},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1090\/gsm\/058"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-540-71050-9"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2020.3028042"},{"key":"ref55","unstructured":"J. Yu , \nR. Lai , \nW. Li , and \nS. Osher , A Fast Proximal Gradient Method and Convergence Analysis for Dynamic Mean Field Planning, preprint, arXiv:2102.13260, 2021."}],"container-title":["SIAM Journal on Scientific Computing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/epubs.siam.org\/doi\/pdf\/10.1137\/22M1490831","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T15:58:59Z","timestamp":1787327939000},"score":1,"resource":{"primary":{"URL":"https:\/\/epubs.siam.org\/doi\/10.1137\/22M1490831"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,8,1]]},"references-count":55,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2023,8,31]]}},"alternative-id":["10.1137\/22M1490831"],"URL":"https:\/\/doi.org\/10.1137\/22m1490831","relation":{},"ISSN":["1064-8275","1095-7197"],"issn-type":[{"value":"1064-8275","type":"print"},{"value":"1095-7197","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,8,1]]}}}