{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T00:02:37Z","timestamp":1791417757214,"version":"4.3.3"},"reference-count":85,"publisher":"Society for Industrial & Applied Mathematics (SIAM)","issue":"1","funder":[{"DOI":"10.13039\/100000185","name":"Defense Advanced Research Projects Agency","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100000185","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["SIAM J. Appl. Dyn. Syst."],"published-print":{"date-parts":[[2020,1]]},"abstract":"<jats:p>The Koopman operator has emerged as a powerful tool for the analysis of nonlinear dynamical systems as it provides coordinate transformations to globally linearize the dynamics. While recent deep learning approaches have been useful in extracting the Koopman operator from a data-driven perspective, several challenges remain. In this work, we formalize the problem of learning the continuous-time Koopman operator with deep neural networks in a measure-theoretic framework. Our approach induces two types of models, differential and recurrent form, the choice of which depends on the availability of the governing equations and data. We then enforce a structural parameterization that renders the realization of the Koopman operator provably stable. A new autoencoder architecture is constructed, such that only the residual of the dynamic mode decomposition is learned. Finally, we employ mean-field variational inference on the aforementioned framework in a hierarchical Bayesian setting to quantify uncertainties in the characterization and prediction of the dynamics of observables. The framework is evaluated on a simple polynomial system, the Duffing oscillator, and an unstable cylinder wake flow with noisy measurements.<\/jats:p>","DOI":"10.1137\/19m1267246","type":"journal-article","created":{"date-parts":[[2020,2,12]],"date-time":"2020-02-12T11:55:20Z","timestamp":1581508520000},"page":"480-509","source":"Crossref","is-referenced-by-count":133,"title":["Physics-Informed Probabilistic Learning of Linear Embeddings of Nonlinear Dynamics with Guaranteed Stability"],"prefix":"10.1137","volume":"19","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2462-362X","authenticated-orcid":true,"given":"Shaowu","family":"Pan","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Karthik","family":"Duraisamy","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"351","published-online":{"date-parts":[[2020,2,12]]},"reference":[{"key":"atypb1","unstructured":"M. Abadi, P. Barham, J. Chen, Z. Chen, A. Davis, J. Dean, M. Devin, S. Ghemawat, G. Irving, M. Isard, et al.\n                      Tensorflow: A system for large-scale machine learning\n                      , in Proceedings of the 12th Symposium on Operating Systems Design and Implementation, 2016, pp. 265-283."},{"key":"atypb2","doi-asserted-by":"publisher","DOI":"10.1137\/17M1125236"},{"key":"atypb3","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevFluids.2.124402"},{"key":"atypb4","unstructured":"D. Arrowsmith and C. M. Place,\n                      Dynamical Systems: Differential Equations, Maps, and Chaotic Behaviour\n                      , Chapman & Hall\/CRC Math. Ser. 5, CRC Press, Boca Raton, FL, 1992."},{"key":"atypb5","doi-asserted-by":"publisher","DOI":"10.1137\/M1124176"},{"key":"atypb6","doi-asserted-by":"publisher","DOI":"10.1080\/01621459.2017.1285773"},{"key":"atypb7","unstructured":"C. Blundell, J. Cornebise, K. Kavukcuoglu, and D. Wierstra,\n                      Weight Uncertainty in Neural Networks\n                      , preprint,https:\/\/arxiv.org\/abs\/1505.05424, 2015."},{"key":"atypb8","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0150171"},{"key":"atypb9","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.1517384113"},{"key":"atypb10","doi-asserted-by":"publisher","DOI":"10.1016\/j.physd.2012.04.006"},{"key":"atypb11","doi-asserted-by":"publisher","DOI":"10.1063\/1.4772195"},{"key":"atypb12","unstructured":"B. Chang, M. Chen, E. Haber, and E. H. Chi,\n                      AntisymmetricRNN: A Dynamical System View on Recurrent Neural Networks\n                      , preprint,https:\/\/arxiv.org\/abs\/1902.09689, 2019."},{"key":"atypb13","doi-asserted-by":"crossref","unstructured":"B. Chang, L. Meng, E. Haber, L. Ruthotto, D. Begert, and E. Holtham,\n                      Reversible architectures for arbitrarily deep residual neural networks\n                      , in Proceedingds of the 32nd AAAI Conference on Artificial Intelligence, 2018.","DOI":"10.1609\/aaai.v32i1.11668"},{"key":"atypb14","doi-asserted-by":"publisher","DOI":"10.1007\/s00332-012-9130-9"},{"key":"atypb15","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.97.7.2968"},{"key":"atypb16","first-page":"853","author":"Denker J. S.","year":"1991","journal-title":"Proceedings of Advances in Neural Information Processing Systems"},{"key":"atypb17","unstructured":"N. B. Erichson, M. Muehlebach, and M. W. Mahoney,\n                      Physics-Informed Autoencoders for Lyapunov-Stable Fluid Flow Prediction\n                      , preprint,https:\/\/arxiv.org\/abs\/1905.10866, 2019."},{"key":"atypb18","doi-asserted-by":"publisher","DOI":"10.1137\/130943637"},{"key":"atypb19","first-page":"515","volume":"1","author":"Gelman A.","year":"2006","journal-title":"Econometrics"},{"key":"atypb20","unstructured":"I. Goodfellow, Y. Bengio, and A. Courville,\n                      Deep Learning\n                      , MIT Press, Cambridge, MA, 2016."},{"key":"atypb21","doi-asserted-by":"publisher","DOI":"10.1098\/rspa.2017.0385"},{"key":"atypb22","first-page":"2348","author":"Graves A.","year":"2011","journal-title":"Proceedings of Advances in Neural Information Processing Systems"},{"key":"atypb23","unstructured":"J. Guckenheimer and P. Holmes,\n                      Nonlinear Oscillations, Dynamical Systems, and Bifurcations of Vector Fields\n                      , Appl. Math. Sci. 42, Springer, New York, 2013."},{"key":"atypb24","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6420\/aa9a90"},{"key":"atypb25","first-page":"770","author":"He K.","year":"2016","journal-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition"},{"key":"atypb26","unstructured":"I. N. Herstein,\n                      Topics in Algebra\n                      , Xerox College Publishing, Lexington, MA, 1975."},{"key":"atypb27","doi-asserted-by":"crossref","unstructured":"G. Hinton and D. Van Camp,\n                      Keeping neural networks simple by minimizing the description length of the weights\n                      , in in Proceedings of the 6th Annual ACM Conference on Computational Learning Theory, 1993.","DOI":"10.1145\/168304.168306"},{"key":"atypb28","doi-asserted-by":"crossref","unstructured":"M. W. Hirsch, S. Smale, and R. L. Devaney,\n                      Differential Equations, Dynamical Systems, and an Introduction to Chaos\n                      , Academic Press, New York, 2012.","DOI":"10.1016\/B978-0-12-382010-5.00015-4"},{"key":"atypb29","first-page":"415","volume":"209","author":"Mercer B. J","year":"1909","journal-title":"Philos. Trans. A"},{"key":"atypb30","first-page":"1","author":"Jasak H.","year":"2007","journal-title":"Dubrovnik Croatia"},{"key":"atypb31","unstructured":"E. Kaiser, J. N. Kutz, and S. L. Brunton,\n                      Data-Driven Discovery of Koopman Eigenfunctions for Control\n                      , preprint,https:\/\/arxiv.org\/abs\/1707.01146, 2017."},{"key":"atypb32","unstructured":"H. K. Khalil,\n                      Noninear Systems\n                      , Prentice-Hall, Englewood Cliffs, NJ, 1996."},{"key":"atypb33","unstructured":"D. P. Kingma and J. Ba,\n                      Adam: A Method for Stochastic Optimization\n                      , preprint,https:\/\/arxiv.org\/abs\/1412.6980, 2014."},{"key":"atypb34","doi-asserted-by":"crossref","unstructured":"A. Klenke,\n                      Probability Theory: A Comprehensive Course\n                      , Springer, New York, 2013.","DOI":"10.1007\/978-1-4471-5361-0"},{"key":"atypb35","unstructured":"S. Klus, P. Koltai, and C. Sch\u00fctte,\n                      On the Numerical Approximation of the Perron-Frobenius and Koopman Operator\n                      , preprint,https:\/\/arxiv.org\/abs\/1512.05997, 2015."},{"key":"atypb36","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.18.3.255"},{"key":"atypb37","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.17.5.315"},{"key":"atypb38","unstructured":"M. Korda and I. Mezi\u0107,\n                      Linear Predictors for Nonlinear Dynamical Systems: Koopman Operator Meets Model Predictive Control\n                      , preprint,https:\/\/arxiv.org\/abs\/1611.03537, 2016."},{"key":"atypb39","doi-asserted-by":"publisher","DOI":"10.1007\/s00332-017-9423-0"},{"key":"atypb40","first-page":"430","volume":"18","author":"Kucukelbir A.","year":"2017","journal-title":"J. Mach. Learn. Res."},{"key":"atypb41","doi-asserted-by":"publisher","DOI":"10.1016\/j.physd.2012.08.017"},{"key":"atypb42","doi-asserted-by":"publisher","DOI":"10.1016\/S0893-6080(05)80131-5"},{"key":"atypb43","doi-asserted-by":"publisher","DOI":"10.1063\/1.4993854"},{"key":"atypb44","first-page":"2378","author":"Liu Q.","year":"2016","journal-title":"Curran Associates"},{"key":"atypb45","first-page":"8854","author":"Liu Q.","year":"2018","journal-title":"Proceedings of Neural Information Processing Systems"},{"key":"atypb46","doi-asserted-by":"publisher","DOI":"10.1137\/18M1187611"},{"key":"atypb47","unstructured":"D. G. Luenberger,\n                      Introduction to Linear and Nonlinear Programming\n                      , Addison-Wesley, Reading, MA, 1973."},{"key":"atypb48","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-018-07210-0"},{"key":"atypb49","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1992.4.3.448"},{"key":"atypb50","doi-asserted-by":"publisher","DOI":"10.1063\/1.4736859"},{"key":"atypb51","first-page":"239","volume":"21","author":"McKay M. D.","year":"1979","journal-title":"Technometrics"},{"key":"atypb52","doi-asserted-by":"crossref","unstructured":"A. Meucci,\n                      Review of Statistical Arbitrage, Cointegration, and Multivariate Ornstein-Uhlenbeck\n                      , SSRN,https:\/\/ssrn.com\/abstract=1404905, 2009.","DOI":"10.2139\/ssrn.1404905"},{"key":"atypb53","doi-asserted-by":"crossref","unstructured":"C. D. Meyer,\n                      Matrix Analysis and Applied Linear Algebra\n                      , SIAM, Philadelphia, 2000.","DOI":"10.1137\/1.9780898719512"},{"key":"atypb54","doi-asserted-by":"publisher","DOI":"10.1007\/s11071-005-2824-x"},{"key":"atypb55","doi-asserted-by":"publisher","DOI":"10.1146\/annurev-fluid-011212-140652"},{"key":"atypb56","first-page":"7034","author":"Mezi\u0107 I.","year":"2015","journal-title":"Proceedings of the IEEE 54th Annual Conference on Decision and Control, IEEE"},{"key":"atypb57","doi-asserted-by":"publisher","DOI":"10.1063\/1.166399"},{"key":"atypb58","doi-asserted-by":"crossref","unstructured":"J. Morton, F. D. Witherden, and M. J. Kochenderfer,\n                      Deep Variational Koopman Models: Inferring Koopman Observations for Uncertainty-Aware Dynamics Modeling and Control\n                      , preprint,https:\/\/arxiv.org\/abs\/1902.09742, 2019.","DOI":"10.24963\/ijcai.2019\/440"},{"key":"atypb59","unstructured":"R. M. Neal,\n                      Bayesian Learning for Neural Networks\n                      , Lecture Notes in Statist. 118, Springer, New York, 2012."},{"key":"atypb60","doi-asserted-by":"publisher","DOI":"10.1137\/18M1177846"},{"key":"atypb61","unstructured":"S. Pan and K. Duraisamy,\n                      Data-Driven Discovery of Closure Models\n                      , preprint,https:\/\/arxiv.org\/abs\/1803.09318, 2018."},{"key":"atypb62","doi-asserted-by":"crossref","unstructured":"S. Pan and K. Duraisamy,\n                      On the Structure of Time-Delay Embedding in Linear Models of Non-Linear Dynamical Systems\n                      , preprint,https:\/\/arxiv.org\/abs\/1902.05198, 2019.","DOI":"10.1063\/5.0010886"},{"key":"atypb63","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcp.2017.07.053"},{"key":"atypb64","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevFluids.2.014604"},{"key":"atypb65","doi-asserted-by":"publisher","DOI":"10.1214\/12-BA730"},{"key":"atypb66","doi-asserted-by":"publisher","DOI":"10.1137\/16M1062296"},{"key":"atypb67","unstructured":"P. Ramachandran, B. Zoph, and Q. V. Le,\n                      Searching for Activation Functions\n                      , preprint,https:\/\/arxiv.org\/abs\/1710.05941, 2017."},{"key":"atypb68","unstructured":"S. M. Ross, J. J. Kelly, R. J. Sullivan, W. J. Perry, D. Mercer, R. M. Davis, T. D. Washburn, E. V. Sager, J. B. Boyce, and V. L. Bristow,\n                      Stochastic Processes\n                      , Vol. 2, Wiley, New York, 1996."},{"key":"atypb69","doi-asserted-by":"publisher","DOI":"10.1017\/S0022112009992059"},{"key":"atypb70","doi-asserted-by":"publisher","DOI":"10.1017\/S0022112010001217"},{"key":"atypb71","doi-asserted-by":"publisher","DOI":"10.1007\/s00162-010-0203-9"},{"key":"atypb72","unstructured":"J. Shi, S. Sun, and J. Zhu,\n                      Kernel implicit variational inference\n                      , in Proceedings of the International Conference on Learning Representations, 2018."},{"key":"atypb73","first-page":"1130","author":"Takeishi N.","year":"2017","journal-title":"Proceedings of Advances in Neural Information Processing Systems"},{"key":"atypb74","unstructured":"D. Tran, A. Kucukelbir, A. B. Dieng, M. Rudolph, D. Liang, and D. M. Blei,\n                      Edward: A Library for Probabilistic Modeling, Inference, and Criticism\n                      , preprint,https:\/\/arxiv.org\/abs\/1610.09787, 2016."},{"key":"atypb75","first-page":"831","author":"Vapnik V.","year":"1992","journal-title":"Proceedings of Advances in Neural Information Processing Systems"},{"key":"atypb76","first-page":"11","author":"Vapnik V. N.","year":"2015","journal-title":"New York"},{"key":"atypb77","doi-asserted-by":"publisher","DOI":"10.1063\/1.5011399"},{"key":"atypb78","unstructured":"S. Wiggins,\n                      Introduction to Applied Nonlinear Dynamical Systems and Chaos\n                      , Vol. 2, Springer, New York, 2003."},{"key":"atypb79","doi-asserted-by":"publisher","DOI":"10.1007\/s00332-015-9258-5"},{"key":"atypb80","unstructured":"M. O. Williams, C. W. Rowley, and I. G. Kevrekidis,\n                      A Kernel-Based Approach to Data-Driven Koopman Spectral Analysis\n                      , preprint,https:\/\/arxiv.org\/abs\/1411.2260, 2014."},{"key":"atypb81","unstructured":"E. Yeung, S. Kundu, and N. Hodas,\n                      Learning Deep Neural Network Representations for Koopman Operators of Nonlinear Dynamical Systems\n                      , preprint,https:\/\/arxiv.org\/abs\/1708.06850, 2017."},{"key":"atypb82","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2018.2889774"},{"key":"atypb83","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcp.2018.04.018"},{"key":"atypb84","doi-asserted-by":"crossref","unstructured":"Y. Zhu, N. Zabaras, P.S. Koutsourelakis, and P. Perdikaris,\n                      Physics-Constrained Deep Learning for High-Dimensional Surrogate Modeling and Uncertainty Quantification Without Labeled Data\n                      , preprint,https:\/\/arxiv.org\/abs\/1901.06314, 2019.","DOI":"10.1016\/j.jcp.2019.05.024"},{"key":"atypb85","unstructured":"J. Zhuo, C. Liu, J. Shi, J. Zhu, N. Chen, and B. Zhang,\n                      Message Passing Stein Variational Gradient Descent\n                      ,https:\/\/arxiv.org\/abs\/1711.04425, 2017."}],"container-title":["SIAM Journal on Applied Dynamical Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/epubs.siam.org\/doi\/pdf\/10.1137\/19M1267246","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,20]],"date-time":"2026-08-20T14:08:38Z","timestamp":1787234918000},"score":1,"resource":{"primary":{"URL":"https:\/\/epubs.siam.org\/doi\/10.1137\/19M1267246"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,1]]},"references-count":85,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2020,1]]}},"alternative-id":["10.1137\/19M1267246"],"URL":"https:\/\/doi.org\/10.1137\/19m1267246","relation":{},"ISSN":["1536-0040"],"issn-type":[{"value":"1536-0040","type":"electronic"}],"subject":[],"published":{"date-parts":[[2020,1]]}}}