{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,3]],"date-time":"2026-04-03T10:27:29Z","timestamp":1775212049063,"version":"3.50.1"},"reference-count":65,"publisher":"Proceedings of the National Academy of Sciences","issue":"31","license":[{"start":{"date-parts":[[2022,1,30]],"date-time":"2022-01-30T00:00:00Z","timestamp":1643500800000},"content-version":"vor","delay-in-days":184,"URL":"https:\/\/www.pnas.org\/site\/aboutpnas\/licenses.xhtml"}],"funder":[{"DOI":"10.13039\/100000181","name":"DOD | USAF | AFMC | Air Force Office of Scientific Research","doi-asserted-by":"publisher","award":["FA9550-18-1-0502"],"award-info":[{"award-number":["FA9550-18-1-0502"]}],"id":[{"id":"10.13039\/100000181","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000181","name":"DOD | USAF | AFMC | Air Force Office of Scientific Research","doi-asserted-by":"publisher","award":["FA9550-18-1-0167"],"award-info":[{"award-number":["FA9550-18-1-0167"]}],"id":[{"id":"10.13039\/100000181","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000006","name":"DOD | United States Navy | Office of Naval Research","doi-asserted-by":"publisher","award":["N00014-18-1-2527"],"award-info":[{"award-number":["N00014-18-1-2527"]}],"id":[{"id":"10.13039\/100000006","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.pnas.org"],"crossmark-restriction":true},"short-container-title":["Proc. Natl. Acad. Sci. U.S.A."],"published-print":{"date-parts":[[2021,8,3]]},"abstract":"<jats:title>Significance<\/jats:title>\n          <jats:p>Mean-field games (MFGs) is an emerging field that models large populations of agents. They play a central role in many disciplines, such as economics, data science, and engineering. Since many applications come in the form of high-dimensional stochastic MFGs, numerical methods that use spatial grids are prone to the curse of dimensionality. To this end, we exploit the variational structure of potential MFGs and reformulate it as a generative adversarial network (GAN) training problem. This reformulation allays a bit the curse of dimensionality when solving high-dimensional MFGs in the stochastic setting, by avoiding spatial grids or uniform sampling in high dimensions, and instead utilizes the structure of the MFG and its connection with GANs.<\/jats:p>","DOI":"10.1073\/pnas.2024713118","type":"journal-article","created":{"date-parts":[[2021,7,30]],"date-time":"2021-07-30T22:30:45Z","timestamp":1627684245000},"update-policy":"https:\/\/doi.org\/10.1073\/pnas.cm10313","source":"Crossref","is-referenced-by-count":57,"title":["Alternating the population and control neural networks to solve high-dimensional stochastic mean-field games"],"prefix":"10.1073","volume":"118","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-2345-3689","authenticated-orcid":false,"given":"Alex Tong","family":"Lin","sequence":"first","affiliation":[{"name":"Department of Mathematics, University of California, Los Angeles, CA 90095;"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2926-4582","authenticated-orcid":false,"given":"Samy Wu","family":"Fung","sequence":"additional","affiliation":[{"name":"Department of Mathematics, University of California, Los Angeles, CA 90095;"},{"name":"Department of Applied Mathematics and Statistics, Colorado School of Mines, Golden, CO 80401;"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2218-5734","authenticated-orcid":false,"given":"Wuchen","family":"Li","sequence":"additional","affiliation":[{"name":"Department of Mathematics, University of South Carolina, Columbia, SC 29208"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6227-0941","authenticated-orcid":false,"given":"Levon","family":"Nurbekyan","sequence":"additional","affiliation":[{"name":"Department of Mathematics, University of California, Los Angeles, CA 90095;"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7900-4658","authenticated-orcid":false,"given":"Stanley J.","family":"Osher","sequence":"additional","affiliation":[{"name":"Department of Mathematics, University of California, Los Angeles, CA 90095;"}]}],"member":"341","published-online":{"date-parts":[[2021,7,30]]},"reference":[{"key":"e_1_3_4_1_2","first-page":"20130397","article-title":"Partial differential equation models in macroeconomics","volume":"372","author":"Achdou Y.","year":"2014","unstructured":"Y. Achdou, F. J. Buera, J.-M. Lasry, P.-L. Lions, B. Moll, Partial differential equation models in macroeconomics. Philos. Trans. R. Soc. Lond. Ser. A Math. Phys. Eng. Sci. 372, 20130397 (2014).","journal-title":"Philos. Trans. R. Soc. Lond. Ser. A Math. Phys. Eng. Sci."},{"key":"e_1_3_4_2_2","doi-asserted-by":"crossref","unstructured":"Y. Achdou J. Han J.-M. Lasry P.-L. Lions B. Moll. \u201cIncome and wealth distribution in macroeconomics: A continuous-time approach.\u201d (NBER Working Paper 23732 National Bureau of Economic Research Cambridge MA 2017). https:\/\/www.nber.org\/papers\/w23732. Accessed 4 July 2021.","DOI":"10.3386\/w23732"},{"key":"e_1_3_4_3_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-14660-2_3"},{"key":"e_1_3_4_4_2","volume-title":"Economic Models and Mean-Field Games Theory","author":"Gomes D. A.","year":"2015","unstructured":"D. A. Gomes, L. Nurbekyan, E. A. Pimentel, Economic Models and Mean-Field Games Theory (IMPA Mathematical Publications, Rio de Janeiro, Brazil, 2015)."},{"key":"e_1_3_4_5_2","doi-asserted-by":"publisher","DOI":"10.1109\/CDC.2017.8263635"},{"key":"e_1_3_4_6_2","doi-asserted-by":"publisher","DOI":"10.1007\/s11579-017-0206-z"},{"key":"e_1_3_4_7_2","first-page":"1","article-title":"Algorithmic trading in competitive markets with mean field games","volume":"52","author":"Casgrain P.","year":"2019","unstructured":"P. Casgrain, S. Jaimungal, Algorithmic trading in competitive markets with mean field games. SIAM News 52, 1\u20132 (2019).","journal-title":"SIAM News"},{"key":"e_1_3_4_8_2","doi-asserted-by":"publisher","DOI":"10.1109\/TSG.2019.2895247"},{"key":"e_1_3_4_9_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.automatica.2018.11.029"},{"key":"e_1_3_4_10_2","unstructured":"D. A. Gomes J. Sa\u00fade A mean-field game approach to price formation in electricity markets. arXiv [Preprint] (2018). https:\/\/arxiv.org\/abs\/1807.07088 (Accessed 4 July 2021)."},{"key":"e_1_3_4_11_2","doi-asserted-by":"publisher","DOI":"10.23919\/ACC.2018.8431807"},{"key":"e_1_3_4_12_2","doi-asserted-by":"publisher","DOI":"10.1088\/1748-3190\/ab49a4"},{"key":"e_1_3_4_13_2","doi-asserted-by":"publisher","DOI":"10.1137\/20M1342690"},{"key":"e_1_3_4_14_2","doi-asserted-by":"publisher","DOI":"10.1080\/17513758.2020.1720322"},{"key":"e_1_3_4_15_2","doi-asserted-by":"publisher","DOI":"10.1007\/s40687-018-0172-y"},{"key":"e_1_3_4_16_2","first-page":"4967","volume-title":"NeurIPS 2019: The 33rd Annual Conference on Neural Information Processing Systems","author":"Guo X.","year":"2019","unstructured":"X. Guo, A. Hu, R. Xu, J. Zhang, \u201cLearning mean-field games\u201d in NeurIPS 2019: The 33rd Annual Conference on Neural Information Processing Systems, H. Wallach , Eds. (Advances in Neural Information Processing Systems, 2019), vol. 32, pp. 4967\u20134977."},{"key":"e_1_3_4_17_2","unstructured":"R. Carmona M. Lauri\u00e8re Z. Tan Linear-quadratic mean-field reinforcement learning: Convergence of policy gradient methods. arXiv [Preprint] (2019). https:\/\/arxiv.org\/abs\/1910.04295 (Accessed 4 July 2021)."},{"key":"e_1_3_4_18_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.crma.2006.09.018"},{"key":"e_1_3_4_19_2","doi-asserted-by":"publisher","DOI":"10.1007\/s11537-007-0657-8"},{"key":"e_1_3_4_20_2","doi-asserted-by":"publisher","DOI":"10.1007\/s10915-017-0436-5"},{"key":"e_1_3_4_21_2","doi-asserted-by":"publisher","DOI":"10.1137\/090758477"},{"key":"e_1_3_4_22_2","doi-asserted-by":"crossref","first-page":"141","DOI":"10.1007\/978-3-319-49996-3_4","volume-title":"Active Particles","author":"Benamou J.-D.","year":"2017","unstructured":"J.-D. Benamou, G. Carlier, F. Santambrogio, \u201cVariational mean field games\u201d in Active Particles, N. Bellomo, P. Degond, E. Tadmor, Eds. (Modeling and Simulation in Science, Engineering, and Technology, Springer, Cham, Switzerland, 2017), vol. 1, pp. 141\u2013171."},{"key":"e_1_3_4_23_2","doi-asserted-by":"publisher","DOI":"10.1007\/s10915-019-00972-9"},{"key":"e_1_3_4_24_2","doi-asserted-by":"publisher","DOI":"10.4310\/AMSA.2018.v3.n2.a1"},{"key":"e_1_3_4_25_2","doi-asserted-by":"publisher","DOI":"10.1137\/18M118640X"},{"key":"e_1_3_4_26_2","doi-asserted-by":"publisher","DOI":"10.1126\/science.153.3731.34"},{"key":"e_1_3_4_27_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-33590-2_7"},{"key":"e_1_3_4_28_2","first-page":"227","article-title":"The variational structure and time-periodic solutions for mean-field games systems","volume":"3","author":"Cirant M.","year":"2018","unstructured":"M. Cirant, L. Nurbekyan, The variational structure and time-periodic solutions for mean-field games systems. Minimax Theory Appl 3, 227\u2013260 (2018).","journal-title":"Minimax Theory Appl"},{"key":"e_1_3_4_29_2","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.1922204117"},{"key":"e_1_3_4_30_2","first-page":"2672","volume-title":"NIPS\u201914: Proceedings of the 27th International Conference on Neural Information Processing Systems","author":"Goodfellow I.","year":"2014","unstructured":"I. Goodfellow , \u201cGenerative adversarial nets\u201d in NIPS\u201914: Proceedings of the 27th International Conference on Neural Information Processing Systems, Z. Ghahramani, M. Welling, C. Cortes, N. D. Lawrence, K. Q. Weinberger, Eds. (MIT Press, Cambridge, MA, 2014), vol. 2, pp. 2672\u20132680."},{"key":"e_1_3_4_31_2","first-page":"214","volume-title":"ICML\u201917: Proceedings of the 34th International Conference on Machine Learning","author":"Arjovsky M.","year":"2017","unstructured":"M. Arjovsky, S. Chintala, L. Bottou, \u201cWasserstein generative adversarial networks\u201d in ICML\u201917: Proceedings of the 34th International Conference on Machine Learning, D. Precup, Y. W. Teh, Eds. (JMLR, 2017), vol. 70, pp. 214\u2013223."},{"key":"e_1_3_4_32_2","first-page":"5767","volume-title":"NIPS\u201917: Proceedings of the 31st International Conference on Neural Information Processing Systems","author":"Gulrajani I.","year":"2017","unstructured":"I. Gulrajani, F. Ahmed, M. Arjovsky, V. Dumoulin, A. C. Courville, \u201cImproved training of Wasserstein GANs\u201d in NIPS\u201917: Proceedings of the 31st International Conference on Neural Information Processing Systems, U. von Luxburg, I. Guyon, S. Bengio, H. Wallach, R. Fergus, Eds. (Curran Associates, Inc., Red Hook, NY, 2017), pp. 5767\u20135777."},{"key":"e_1_3_4_33_2","unstructured":"A. T. Lin W. Li S. Osher G. Mont\u00fafar Wasserstein proximal of GANs. UCLA CAM [Preprint] (2018). ftp:\/\/ftp.math.ucla.edu\/pub\/camreport\/cam18-53.pdf (Accessed 4 July 2021)."},{"key":"e_1_3_4_34_2","first-page":"1716","volume-title":"Proceedings of the 36th International Conference on Machine Learning","author":"Dukler Y.","year":"2019","unstructured":"Y. Dukler, W. Li, A. Lin, G. Montufar, \u201cWasserstein of Wasserstein loss for learning generative models\u201d in Proceedings of the 36th International Conference on Machine Learning, K. Chaudhuri, R. Salakhutdinov, Eds. (Proceedings of Machine Learning Research, PMLR, 2019), vol. 97, pp. 1716\u20131725."},{"key":"e_1_3_4_35_2","unstructured":"K. F. C. Chu K. Minami Smoothness and stability in GANs. arXiv [Preprint] (2020). https:\/\/arxiv.org\/abs\/2002.04185 (Accessed 4 July 2021)."},{"key":"e_1_3_4_36_2","doi-asserted-by":"publisher","DOI":"10.1051\/cocv\/2014044"},{"key":"e_1_3_4_37_2","doi-asserted-by":"publisher","DOI":"10.1007\/s00030-015-0323-4"},{"key":"e_1_3_4_38_2","unstructured":"H. Cao X. Guo M. Lauri\u00e8re Connecting GANs and MFGs. arXiv [Preprint] (2020). https:\/\/arxiv.org\/abs\/2002.04112 (Accessed 4 July 2021)."},{"key":"e_1_3_4_39_2","author":"Villani C.","year":"2003","unstructured":"C. Villani, Topics in Optimal Transportation (Graduate Studies in Mathematics, American Mathematical Society, Providence, RI, 2003), vol. 58.","journal-title":"Topics in Optimal Transportation"},{"key":"e_1_3_4_40_2","unstructured":"D. Onken S. W. Fung X. Li L. Ruthotto OT-Flow: Fast and accurate continuous normalizing flows via optimal transport. arXiv [Preprint] (2020). https:\/\/arxiv.org\/abs\/2006.00104 (Accessed 4 July 2021)."},{"key":"e_1_3_4_41_2","unstructured":"T. Q. Chen Y. Rubanova J. Bettencourt D. K. Duvenaud \u201cNeural ordinary differential equations\u201d in NIPS\u201918: Proceedings of the 32nd International Conference on Neural Information Processing Systems S. Bengio H. M. Wallach H. Larochelle K. Grauman N. Cesa-Bianchi Eds. (Curran Associates Red Hook NY 2018) pp. 6572\u20136583."},{"key":"e_1_3_4_42_2","doi-asserted-by":"publisher","DOI":"10.4310\/CMS.2018.v16.n7.a9"},{"key":"e_1_3_4_43_2","doi-asserted-by":"publisher","DOI":"10.1186\/s40687-016-0068-7"},{"key":"e_1_3_4_44_2","first-page":"9649","volume-title":"NIPS\u201918: Proceedings of the 32nd International Conference on Neural Information Processing Systems","author":"Wai H.-T.","year":"2018","unstructured":"H.-T. Wai, Z. Yang, Z. Wang, M. Hong, \u201cMulti-agent reinforcement learning via double averaging primal-dual optimization\u201d in NIPS\u201918: Proceedings of the 32nd International Conference on Neural Information Processing Systems, S. Bengio, H. M. Wallach, H. Larochelle, K. Grauman, N. Cesa-Bianchi, Eds. (Curran Associates, Red Hook, NY, 2018), pp. 9649\u20139660."},{"key":"e_1_3_4_45_2","unstructured":"X. Guo A. Hu R. Xu J. Zhang A general framework for learning mean-field games. arXiv [Preprint] (2020). https:\/\/arxiv.org\/abs\/2003.06069 (Accessed 4 July 2021)."},{"key":"e_1_3_4_46_2","volume-title":"6th International Conference on Learning Representations","author":"Yang J.","year":"2018","unstructured":"J. Yang, X. Ye, R. Trivedi, H. Xu, H. Zha, \u201cDeep mean field games for learning optimal behavior policy of large populations\u201d in 6th International Conference on Learning Representations (ICLR, 2018)."},{"key":"e_1_3_4_47_2","volume-title":"8th International Conference on Learning Representations","author":"Fu Z.","year":"2020","unstructured":"Z. Fu, Z. Yang, Y. Chen, Z. Wang, \u201cActor-critic provably finds Nash equilibria of linear-quadratic mean-field games\u201d in 8th International Conference on Learning Representations (ICLR, 2020)."},{"key":"e_1_3_4_48_2","first-page":"1608","volume-title":"Proceedings of the Twenty-First International Conference on Artificial Intelligence and Statistics","author":"Genevay A.","year":"2018","unstructured":"A. Genevay, G. Peyre, M. Cuturi, \u201cLearning generative models with Sinkhorn divergences\u201d in Proceedings of the Twenty-First International Conference on Artificial Intelligence and Statistics, A. Storkey, F. Perez-Cruz, Eds. (Proceedings of Machine Learning Research, PMLR, 2018), vol. 84, pp. 1608\u20131617."},{"key":"e_1_3_4_49_2","volume-title":"6th International Conference on Learning Representations","author":"Salimans T.","year":"2018","unstructured":"T. Salimans, H. Zhang, A. Radford, D. Metaxas, \u201cImproving GANs using optimal transport\u201d in 6th International Conference on Learning Representations (ICLR, 2018)."},{"key":"e_1_3_4_50_2","first-page":"7091","volume-title":"NIPS\u201918: Proceedings of the 32nd International Conference on Neural Information Processing Systems","author":"Sanjabi M.","year":"2018","unstructured":"M. Sanjabi, J. Ba, M. Razaviyayn, J. D. Lee, \u201cOn the convergence and robustness of training GANs with regularized optimal transport\u201d in NIPS\u201918: Proceedings of the 32nd International Conference on Neural Information Processing Systems, S. Bengio, H. M. Wallach, H. Larochelle, K. Grauman, N. Cesa-Bianchi, Eds. (Curran Associates, Red Hook, NY, 2018), pp. 7091\u20137101."},{"key":"e_1_3_4_51_2","first-page":"6813","volume-title":"NeurIPS 2019: The 33rd Annual Conference on Neural Information Processing Systems","author":"Tanaka A.","year":"2019","unstructured":"A. Tanaka, \u201cDiscriminator optimal transport\u201d in NeurIPS 2019: The 33rd Annual Conference on Neural Information Processing Systems, H. Wallach , Eds. (Advances in Neural Information Processing Systems, 2019), vol. 32, pp. 6813\u20136823."},{"key":"e_1_3_4_52_2","unstructured":"J. Lin K. Lensink E. Haber Fluid flow mass transport for generative networks. arXiv [Preprint] (2019). https:\/\/arxiv.org\/abs\/1910.01694 (Accessed 4 July 2021)."},{"key":"e_1_3_4_53_2","doi-asserted-by":"publisher","DOI":"10.1561\/2200000073"},{"key":"e_1_3_4_54_2","unstructured":"C. Finlay B.-H. Jacobsen L. Nurbekyan A. M. Oberman How to train your neural ODE. arXiv [Preprint] (2020). https:\/\/arxiv.org\/abs\/2002.02798 (Accessed 4 July 2021)."},{"key":"e_1_3_4_55_2","volume-title":"International Conference on Learning Representations 2019","author":"Grathwohl W.","year":"2019","unstructured":"W. Grathwohl, R. T. Q. Chen, J. Betterncourt, I. Sutskever, D. Duvenaud, \u201cFFJORD: Free-form continuous dynamics for scalable reversible generative models\u201d in International Conference on Learning Representations 2019 (ICLR, 2019)."},{"key":"e_1_3_4_56_2","unstructured":"D. Onken L. Ruthotto Discretize-optimize vs. optimize-discretize for time-series regression and continuous normalizing flows. arXiv [Preprint] (2020). https:\/\/arxiv.org\/abs\/2005.13420 (Accessed 4 July 2021)."},{"key":"e_1_3_4_57_2","doi-asserted-by":"crossref","unstructured":"C. Parkinson D. Arnold A. L. Bertozzi S. Osher A model for optimal human navigation with stochastic effects. arXiv [Preprint] (2020). https:\/\/arxiv.org\/abs\/2005.03615 (Accessed 4 July 2021).","DOI":"10.1137\/19M1296537"},{"key":"e_1_3_4_58_2","doi-asserted-by":"publisher","DOI":"10.1214\/aoms\/1177728190"},{"key":"e_1_3_4_59_2","doi-asserted-by":"publisher","DOI":"10.1214\/aoms\/1177704472"},{"key":"e_1_3_4_60_2","doi-asserted-by":"crossref","unstructured":"S. Liu L. Nurbekyan Splitting methods for a class of non-potential mean field games. arXiv [Preprint] (2020). https:\/\/arxiv.org\/abs\/2007.00099 (Accessed 4 July 2021).","DOI":"10.3934\/jdg.2021014"},{"key":"e_1_3_4_61_2","unstructured":"H. Cao X. Guo Approximation and convergence of GANs training: An SDE approach. arXiv [Preprint] (2020). https:\/\/arxiv.org\/abs\/2006.02047 (Accessed 4 July 2021)."},{"key":"e_1_3_4_62_2","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6420\/aa9a90"},{"key":"e_1_3_4_63_2","first-page":"1","article-title":"Deep neural networks motivated by partial differential equations","volume":"62","author":"Ruthotto L.","year":"2019","unstructured":"L. Ruthotto, E. Haber, Deep neural networks motivated by partial differential equations. J. Math. Imag. Vis. 62, 1\u201313 (2019).","journal-title":"J. Math. Imag. Vis."},{"key":"e_1_3_4_64_2","first-page":"291","volume-title":"International Conference on Machine Learning","author":"Anil C.","year":"2019","unstructured":"C. Anil, J. Lucas, R. Grosse, \u201cSorting out Lipschitz function approximation\u201d in International Conference on Machine Learning, K. Chaudhuri, R. Salakhutdinov, Eds. (Proceedings of Machine Learning Research, PMLR, 2019), pp. 291\u2013301."},{"key":"e_1_3_4_65_2","doi-asserted-by":"publisher","DOI":"10.1016\/S0169-7161(04)24009-3"}],"container-title":["Proceedings of the National Academy of Sciences"],"original-title":[],"language":"en","link":[{"URL":"http:\/\/www.pnas.org\/syndication\/doi\/10.1073\/pnas.2024713118","content-type":"unspecified","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pnas.org\/doi\/pdf\/10.1073\/pnas.2024713118","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,4,13]],"date-time":"2022-04-13T10:59:08Z","timestamp":1649847548000},"score":1,"resource":{"primary":{"URL":"https:\/\/pnas.org\/doi\/full\/10.1073\/pnas.2024713118"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,7,30]]},"references-count":65,"journal-issue":{"issue":"31","published-print":{"date-parts":[[2021,8,3]]}},"alternative-id":["10.1073\/pnas.2024713118"],"URL":"https:\/\/doi.org\/10.1073\/pnas.2024713118","relation":{},"ISSN":["0027-8424","1091-6490"],"issn-type":[{"value":"0027-8424","type":"print"},{"value":"1091-6490","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,7,30]]},"assertion":[{"value":"2021-07-30","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e2024713118"}}