{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,26]],"date-time":"2026-03-26T07:50:09Z","timestamp":1774511409202,"version":"3.50.1"},"reference-count":32,"publisher":"MDPI AG","issue":"10","license":[{"start":{"date-parts":[[2019,9,23]],"date-time":"2019-09-23T00:00:00Z","timestamp":1569196800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100000183","name":"Army research office","doi-asserted-by":"publisher","award":["W911NF-17-1-0306"],"award-info":[{"award-number":["W911NF-17-1-0306"]}],"id":[{"id":"10.13039\/100000183","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000006","name":"Office of Naval Research","doi-asserted-by":"publisher","award":["N00014-15-1-2381"],"award-info":[{"award-number":["N00014-15-1-2381"]}],"id":[{"id":"10.13039\/100000006","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Entropy"],"abstract":"<jats:p>The ability to characterize and predict extreme events is a vital topic in fields ranging from finance to ocean engineering. Typically, the most-extreme events are also the most-rare, and it is this property that makes data collection and direct simulation challenging. We consider the problem of deriving optimal predictors of extremes directly from data characterizing a complex system, by formulating the problem in the context of binary classification. Specifically, we assume that a training dataset consists of: (i) indicator time series specifying on whether or not an extreme event occurs; and (ii) observables time series, which are employed to formulate efficient predictors. We employ and assess standard binary classification criteria for the selection of optimal predictors, such as total and balanced error and area under the curve, in the context of extreme event prediction. For physical systems for which there is sufficient separation between the extreme and regular events, i.e., extremes are distinguishably larger compared with regular events, we prove the existence of optimal extreme event thresholds that lead to efficient predictors. Moreover, motivated by the special character of extreme events, i.e., the very low rate of occurrence, we formulate a new objective function for the selection of predictors. This objective is constructed from the same principles as receiver operating characteristic curves, and exhibits a geometric connection to the regime separation property. We demonstrate the application of the new selection criterion to the advance prediction of intermittent extreme events in two challenging complex systems: the Majda\u2013McLaughlin\u2013Tabak model, a 1D nonlinear, dispersive wave model, and the 2D Kolmogorov flow model, which exhibits extreme dissipation events.<\/jats:p>","DOI":"10.3390\/e21100925","type":"journal-article","created":{"date-parts":[[2019,9,23]],"date-time":"2019-09-23T11:02:00Z","timestamp":1569236520000},"page":"925","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":39,"title":["Machine Learning Predictors of Extreme Events Occurring in Complex Dynamical Systems"],"prefix":"10.3390","volume":"21","author":[{"given":"Stephen","family":"Guth","sequence":"first","affiliation":[{"name":"Department of Mechanical Engineering, Massachusetts Institute of Technology, 77 Massachusetts Ave., Cambridge, MA 02139, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0302-0691","authenticated-orcid":false,"given":"Themistoklis P.","family":"Sapsis","sequence":"additional","affiliation":[{"name":"Department of Mechanical Engineering, Massachusetts Institute of Technology, 77 Massachusetts Ave., Cambridge, MA 02139, USA"}]}],"member":"1968","published-online":{"date-parts":[[2019,9,23]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"273","DOI":"10.1214\/ss\/1177009939","article-title":"Bayesian Experimental Design: A Review","volume":"10","author":"Chaloner","year":"1995","journal-title":"Stat. Sci."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"525","DOI":"10.1137\/S0036144500378302","article-title":"An Algorithmic Introduction to Numerical Simulation of Stochastic Differential Equations","volume":"43","author":"Higham","year":"2001","journal-title":"SIAM Rev."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"287","DOI":"10.1146\/annurev.fluid.40.111406.102203","article-title":"Oceanic Rogue Waves","volume":"40","author":"Dysthe","year":"2008","journal-title":"Annu. Rev. Fluid Mech."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Kharif, C., Pelinovsky, E., and Slunyaev, A. (2009). Rogue Waves in the Ocean, Observation, Theories and Modeling. Advances in Geophysical and Environmental Mechanics and Mathematics Series, Springer.","DOI":"10.1007\/978-3-540-88419-4_2"},{"key":"ref_5","unstructured":"Li, F. (2017). Modelling the Stock Market Using a Multi-Scale Approach. [Master\u2019s Thesis, University of Leicester]."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1371\/journal.pone.0000049","article-title":"Adaptive Response of a Gene Network to Environmental Changes by Fitness-Induced Attractor Selection","volume":"1","author":"Kashiwagi","year":"2006","journal-title":"PLoS ONE"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"580","DOI":"10.1016\/j.nucengdes.2008.11.005","article-title":"Estimation of the functional failure probability of a thermal-hydraulic passive system by Subset Simulation","volume":"239","author":"Zio","year":"2009","journal-title":"Nucl. Eng. Des."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"478","DOI":"10.1016\/j.enconman.2015.11.032","article-title":"Hamiltonian modeling of multi-hydro-turbine governing systems with sharing common penstock and dynamic analyses under shock load","volume":"108","author":"Beibei","year":"2016","journal-title":"Energy Convers. Manag."},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Varadhan, S.R.S. (1984). Large Deviations and Applications, SIAM.","DOI":"10.1137\/1.9781611970241"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"391","DOI":"10.1146\/annurev.physchem.040808.090412","article-title":"Transition-path theory and path-finding algorithms for the study of rare events","volume":"61","author":"E","year":"2010","journal-title":"Annu. Rev. Phys. Chem."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"1687","DOI":"10.4310\/CMS.2016.v14.n6.a11","article-title":"Predicting Fat-Tailed Intermittent Probability Distributions in Passive Scalar Turbulence with Imperfect Models through Empirical Information Theory","volume":"14","author":"Qi","year":"2016","journal-title":"Commun. Math. Sci."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"709","DOI":"10.1137\/140978235","article-title":"Probabilistic Description of Extreme Events in Intermittently Unstable Dynamical Systems Excited by Correlated Stochastic Processes","volume":"3","author":"Mohamad","year":"2015","journal-title":"SIAM\/ASA J. Uncertain. Quantif."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"3982","DOI":"10.1073\/pnas.1820467116","article-title":"Statistical dynamical model to predict extreme events and anomalous features in shallow water waves with abrupt depth change","volume":"116","author":"Majda","year":"2018","journal-title":"Proc. Natl. Acad. Sci. USA"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"e1701533","DOI":"10.1126\/sciadv.1701533","article-title":"A variational approach to probing extreme events in turbulent dynamical systems","volume":"3","author":"Farazmand","year":"2017","journal-title":"Sci. Adv."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Wan, Z.Y., Vlachas, P.R., Koumoutsakos, P., and Sapsis, T.P. (2018). Data-assisted reduced-order modeling of extreme events in complex dynamical systems. PLoS ONE.","DOI":"10.1371\/journal.pone.0197704"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"11138","DOI":"10.1073\/pnas.1813263115","article-title":"Sequential sampling strategy for extreme event statistics in nonlinear dynamical systems","volume":"115","author":"Mohamad","year":"2018","journal-title":"Proc. Natl. Acad. Sci. USA"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"508","DOI":"10.1186\/cc3000","article-title":"Statistics review 13: Receiver operating characteristic curves","volume":"8","author":"Ball","year":"2004","journal-title":"Crit. Care"},{"key":"ref_18","first-page":"1263","article-title":"Learning from imbalanced data","volume":"29","author":"He","year":"2009","journal-title":"IEEE Trans. Knowl. Data Eng."},{"key":"ref_19","first-page":"1","article-title":"The Precision-Recall Plot Is More Informative than the ROC Plot When Evaluating Binary Classifiers on Imbalanced Datasets","volume":"10","year":"2015","journal-title":"PLoS ONE"},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Cousins, W., and Sapsis, T.P. (2016). Reduced-order precursors of rare events in unidirectional nonlinear water waves. J. Fluid Mech., 790.","DOI":"10.1017\/jfm.2016.13"},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"418","DOI":"10.1016\/j.jcp.2017.03.054","article-title":"Reduced-order prediction of rogue waves in two-dimensional deep-water waves","volume":"340","author":"Farazmand","year":"2017","journal-title":"J. Comput. Phys."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"855","DOI":"10.1073\/pnas.1710670115","article-title":"Rogue Waves and Large Deviations in Deep Sea","volume":"115","author":"Dematteis","year":"2018","journal-title":"Proc. Natl. Acad. Sci. USA"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"9","DOI":"10.1007\/BF02679124","article-title":"A one-dimensional model for dispersive wave turbulence","volume":"7","author":"Majda","year":"1997","journal-title":"J. Nonlinear Sci."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"681","DOI":"10.1063\/1.858074","article-title":"An investigation of chaotic Kolmogorov flows","volume":"3","author":"Platt","year":"1991","journal-title":"Phys. Fluids A"},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"48","DOI":"10.1016\/j.physd.2014.04.012","article-title":"Quantification and prediction of extreme events in a one-dimensional nonlinear dispersive wave model","volume":"280","author":"Cousins","year":"2014","journal-title":"Physica D"},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Mohamad, M., and Sapsis, T. (2016). Probabilistic response and rare events in Mathieu\u2019s equation under correlated parametric excitation. Ocean Eng., 120.","DOI":"10.1016\/j.oceaneng.2016.03.008"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"288","DOI":"10.1016\/j.jcp.2016.06.047","article-title":"A probabilistic decomposition-synthesis method for the quantification of rare events due to internal instabilities","volume":"322","author":"Mohamad","year":"2016","journal-title":"J. Comput. Phys."},{"key":"ref_28","first-page":"551","article-title":"Dispersive wave turbulence in one dimension","volume":"152\u2013153","author":"David","year":"2001","journal-title":"Physica D"},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Benno Rumpf, L.B. (2005). Weak turbulence and collapses in the Majda\u2013McLaughlin\u2013Tabak equation: Fluxes in wavenumber and in amplitude space. Physica D, 188\u2013203.","DOI":"10.1016\/j.physd.2005.04.012"},{"key":"ref_30","first-page":"429","article-title":"Theory of Communication","volume":"93","author":"Gabor","year":"1946","journal-title":"J. Inst. Electr. Eng."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"3982","DOI":"10.1039\/C4EE02841D","article-title":"Solid\u2013liquid separation by particle-flow-instability","volume":"7","author":"Wang","year":"2014","journal-title":"Energy Environ. Sci."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Vu, K.K., D\u2019Ambrosio, C., Hamadi, Y., and Liberti, L. (2016). Surrogate-based methods for black-box optimization. Int. Trans. Oper. Res., 24.","DOI":"10.1111\/itor.12292"}],"container-title":["Entropy"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1099-4300\/21\/10\/925\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T13:23:23Z","timestamp":1760189003000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1099-4300\/21\/10\/925"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2019,9,23]]},"references-count":32,"journal-issue":{"issue":"10","published-online":{"date-parts":[[2019,10]]}},"alternative-id":["e21100925"],"URL":"https:\/\/doi.org\/10.3390\/e21100925","relation":{},"ISSN":["1099-4300"],"issn-type":[{"value":"1099-4300","type":"electronic"}],"subject":[],"published":{"date-parts":[[2019,9,23]]}}}