{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T23:16:23Z","timestamp":1790723783471,"version":"4.1.0"},"reference-count":59,"publisher":"Society for Industrial & Applied Mathematics (SIAM)","issue":"1","funder":[{"DOI":"10.13039\/501100000780","name":"European Union","doi-asserted-by":"crossref","award":["813261"],"award-info":[{"award-number":["813261"]}],"id":[{"id":"10.13039\/501100000780","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["SIAM J. Sci. Comput."],"published-print":{"date-parts":[[2023,2,28]]},"abstract":"<jats:p>Abstract.<\/jats:p>\n                  <jats:p>In this paper, we propose a deep learning based numerical scheme for strongly coupled forward backward stochastic differential equations (FBSDEs), stemming from stochastic control. It is a modification of the deep BSDE method in which the initial value to the backward equation is not a free parameter, and with a new loss function being the weighted sum of the cost of the control problem, and a variance term which coincides with the mean squared error in the terminal condition. We show by a numerical example that a direct extension of the classical deep BSDE method to FBSDEs fails for a simple linear-quadratic control problem, and we motivate why the new method works. Under regularity and boundedness assumptions on the exact controls of time continuous and time discrete control problems, we provide an error analysis for our method. We show empirically that the method converges for three different problems, one being the one that failed for a direct extension of the deep BSDE method.<\/jats:p>","DOI":"10.1137\/22m1478057","type":"journal-article","created":{"date-parts":[[2023,2,24]],"date-time":"2023-02-24T14:40:31Z","timestamp":1677249631000},"page":"A226-A255","source":"Crossref","is-referenced-by-count":20,"title":["Convergence of a Robust Deep FBSDE Method for Stochastic Control"],"prefix":"10.1137","volume":"45","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7186-4840","authenticated-orcid":true,"given":"Kristoffer","family":"Andersson","sequence":"first","affiliation":[{"name":"Research Group of Scientific Computing, Centrum Wiskunde & Informatica, Amsterdam, Netherlands."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8342-4997","authenticated-orcid":true,"given":"Adam","family":"Andersson","sequence":"additional","affiliation":[{"name":"Research Group of Computational Mathematics, Chalmers University of Technology and the University of Gothenburg, and Saab AB Radar Solutions, Gothenburg, Sweden."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"C. W.","family":"Oosterlee","sequence":"additional","affiliation":[{"name":"Mathematical Institute, Utrecht University, Utrecht, Netherlands."}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"351","published-online":{"date-parts":[[2023,2,23]]},"reference":[{"key":"ref1","unstructured":"K. Andersson , Approximate Stochastic Control Based on Deep Learning and Forward Backward Stochastic Differential Equations, Master\u2019s thesis, Chalmers University of Technology, 2019."},{"key":"ref2","doi-asserted-by":"crossref","unstructured":"F. Antonelli  , Backward Forward Stochastic Differential Equations, Ph.D. thesis, Purdue University, 1993.","DOI":"10.1214\/aoap\/1177005363"},{"key":"ref3","volume-title":"Introduction to Stochastic Control Theory","author":"\u00c5str\u00f6m K. J.","year":"2012"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1137\/19M1297919"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1007\/s10915-021-01590-0"},{"key":"ref6","unstructured":"C. Beck , \nM. Hutzenthaler , \nA. Jentzen , and \nB. Kuckuck , An Overview on Deep Learning-Based Approximation Methods for Partial Differential Equations, preprint, https:\/\/arxiv.org\/abs\/2012.12348, 2020."},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1007\/s00332-018-9525-3"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1016\/j.spa.2007.03.005"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-25746-9_8"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1137\/19M125649X"},{"key":"ref11","doi-asserted-by":"crossref","first-page":"389","DOI":"10.3934\/dcdsb.2010.14.389","volume":"14","author":"Beyn W.-J.","year":"2010","journal-title":"Discrete Contin. Dyn. Syst. Ser. B"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1051\/cocv\/2018045"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1016\/j.spa.2004.01.001"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1007\/BF00935345"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1007\/s10915-019-00908-3"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1080\/00207160.2019.1658868"},{"key":"ref17","unstructured":"J. Chessari  and \nR. Kawai , Numerical Methods for Backward Stochastic Differential Equations: A Survey, preprint, https:\/\/arxiv.org\/abs\/2101.08936, 2021."},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1016\/j.spa.2010.03.015"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1007\/BF02551274"},{"key":"ref20","unstructured":"B. Dai , \nV. R. Surabhi , \nP. Krishnamurthy , and \nF. Khorrami , Learning Locomotion Controllers for Walking Using Deep FBSDE, preprint, https:\/\/arxiv.org\/abs\/2107.07931, 2021."},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1007\/s00365-021-09541-6"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1214\/10-AAP723"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1137\/080718061"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1007\/s10690-019-09271-7"},{"key":"ref25","doi-asserted-by":"crossref","unstructured":"A. Gnoatto , \nC. Reisinger , and \nA. Picarelli , Deep xVA Solver\u2014A Neural Network Based Counterparty Credit Risk Management Framework, SSRN 3594076, 2020.","DOI":"10.2139\/ssrn.3594076"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1016\/j.spa.2006.10.007"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1137\/090755060"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1137\/16M106371X"},{"key":"ref29","unstructured":"P. Grohs , \nF. Hornung , \nA. Jentzen , and \nP. von Wurstemberger , A Proof that Artificial Neural Networks Overcome the Curse of Dimensionality in the Numerical Approximation of Black\u2013Scholes Partial Differential Equations, https:\/\/arxiv.org\/abs\/1809.02362, 2018."},{"key":"ref30","unstructured":"J. Han , Deep Learning Approximation for Stochastic Control Problems, preprint, https:\/\/arxiv.org\/abs\/1611.07422, 2016."},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.1718942115"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1186\/s41546-020-00047-w"},{"key":"ref33","doi-asserted-by":"crossref","unstructured":"P. Henry-Labordere , Deep Primal-Dual Algorithm for BSDEs: Applications of Machine Learning to CVA and IM, SSRN 3071506, 2017.","DOI":"10.2139\/ssrn.3071506"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1016\/j.cam.2015.10.019"},{"key":"ref35","unstructured":"C. Hur\u00e9 , \nH. Pham , and \nX. Warin , Some Machine Learning Schemes for High-Dimensional Nonlinear PDEs, preprint, https:\/\/arxiv.org\/abs\/1902.01599, 2019."},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1090\/mcom\/3514"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1007\/s10915-018-00903-0"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1007\/s42985-019-0006-9"},{"key":"ref39","unstructured":"A. Jentzen , \nD. Salimova , and \nT. Welti , A Proof that Deep Artificial Neural Networks Overcome the Curse of Dimensionality in the Numerical Approximation of Kolmogorov Partial Differential Equations with Constant Diffusion and Nonlinear Drift Coefficients, preprint, https:\/\/arxiv.org\/abs\/1809.07321, 2018."},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/MIS.2020.2971597"},{"key":"ref41","unstructured":"S. Ji , \nS. Peng , \nY. Peng , and \nX. Zhang , A Control Method for Solving High-Dimensional Hamiltonian Systems Through Deep Neural Networks, preprint, https:\/\/arxiv.org\/abs\/2111.02636, 2021."},{"key":"ref42","doi-asserted-by":"crossref","unstructured":"Y. Jiang  and \nJ. Li , Convergence of the Deep BSDE Method for FBSDEs with Non-Lipschitz Coefficients, preprint, https:\/\/arxiv.org\/abs\/2101.01869, 2021.","DOI":"10.3934\/puqr.2021019"},{"key":"ref43","unstructured":"L. Kapllani  and \nL. Teng , Deep Learning Algorithms for Solving High Dimensional Nonlinear Backward Stochastic Differential Equations, preprint, https:\/\/arxiv.org\/abs\/2010.01319, 2020."},{"key":"ref44","unstructured":"D. P. Kingma  and \nJ. Ba , Adam: A Method for Stochastic Optimization, preprint, https:\/\/arxiv.org\/abs\/1412.6980, 2014."},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1016\/j.cam.2017.05.015"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1007\/s10543-011-0341-5"},{"key":"ref47","doi-asserted-by":"crossref","unstructured":"Y. Liu , \nY. Wang , \nZ. Zhuang , and \nX. Guo , Deep FBSDE Controller for Attitude Control of Hypersonic Aircraft, in Proceedings of the 6th IEEE International Conference on Advanced Robotics and Mechatronics, IEEE, 2021, pp. 294\u2013299.","DOI":"10.1109\/ICARM52023.2021.9536107"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1214\/aoap\/1015961165"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1007\/BF01192258"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1002\/mana.19911510114"},{"key":"ref51","unstructured":"B. Negyesi , \nK. Andersson , and \nC. W. Oosterlee , The One Step Malliavin scheme: New Discretization of BSDEs Implemented with Deep Learning Regressions, preprint, https:\/\/arxiv.org\/abs\/2110.05421, 2021."},{"key":"ref52","unstructured":"M. Pereira , \nZ. Wang , \nI. Exarchos , and \nE. A. Theodorou , Learning Deep Stochastic Optimal Control Policies Using Forward-Backward SDEs, preprint, https:\/\/arxiv.org\/abs\/1902.03986, 2019."},{"key":"ref53","unstructured":"M. Raissi , Forward-Backward Stochastic Neural Networks: Deep Learning of High-Dimensional Partial Differential Equations, preprint, https:\/\/arxiv.org\/abs\/1804.07010, 2018."},{"key":"ref54","series-title":"Grundlehren Math. Wiss. 317","volume-title":"Variational Analysis","author":"Rockafellar R. T.","year":"2009"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1137\/130913183"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1016\/j.apnum.2015.12.003"},{"key":"ref57","first-page":"125","volume":"25","author":"Teng L.","year":"2021","journal-title":"J. Comput. Finance"},{"key":"ref58","unstructured":"Y. Wang  and \nY.H. Ni , Deep BSDE-ML Learning and Its Application to Model-Free Optimal Control, preprint, https:\/\/arxiv.org\/abs\/2201.01318, 2022."},{"key":"ref59","doi-asserted-by":"crossref","unstructured":"Y. Yu , \nB. Hientzsch  and \nN. Ganesan , Backward Deep BSDE Methods and Applications to Nonlinear Problems, preprint, https:\/\/arxiv.org\/abs\/2006.07635, 2020.","DOI":"10.2139\/ssrn.3626208"}],"container-title":["SIAM Journal on Scientific Computing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/epubs.siam.org\/doi\/pdf\/10.1137\/22M1478057","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T15:39:07Z","timestamp":1787326747000},"score":1,"resource":{"primary":{"URL":"https:\/\/epubs.siam.org\/doi\/10.1137\/22M1478057"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,2,23]]},"references-count":59,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2023,2,28]]}},"alternative-id":["10.1137\/22M1478057"],"URL":"https:\/\/doi.org\/10.1137\/22m1478057","relation":{},"ISSN":["1064-8275","1095-7197"],"issn-type":[{"value":"1064-8275","type":"print"},{"value":"1095-7197","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,2,23]]}}}