{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,9]],"date-time":"2026-03-09T00:18:39Z","timestamp":1773015519747,"version":"3.50.1"},"reference-count":47,"publisher":"Wiley","issue":"3","license":[{"start":{"date-parts":[[2026,1,28]],"date-time":"2026-01-28T00:00:00Z","timestamp":1769558400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2026,1,28]],"date-time":"2026-01-28T00:00:00Z","timestamp":1769558400000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"funder":[{"DOI":"10.13039\/501100000266","name":"Engineering and Physical Sciences Research Council","doi-asserted-by":"publisher","award":["EP\/X018288\/1"],"award-info":[{"award-number":["EP\/X018288\/1"]}],"id":[{"id":"10.13039\/501100000266","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Expert Systems"],"published-print":{"date-parts":[[2026,3]]},"abstract":"<jats:title>ABSTRACT<\/jats:title>\n                  <jats:p>This work proposes the use of conditional flow\u2010based generative models to learn an approximation of the distribution of the critical points of a cost function. This approximation is used to incrementally identify all critical points, in the feasible domain of said function, by iteratively alternating the sampling of the distribution and the retraining of the model with the newly discovered points. This paper will focus, in particular, on the identification and conditional generation of all local minima in the case in which the value of the cost function is subject to some uncertain parameters. The target application is the study of complex dynamical systems. It will be shown that when the cost function represents the potential of a dynamical system, the proposed flow\u2010based model can be used to generate minima conditional to their degree of stability or metastability. In dynamical systems subject to uncertainty in the dynamics, the existence of the minima and their stability characteristics are a function of the uncertain parameters. Thus, the proposed model architecture incorporates a conditional variable that can be the value of the uncertain parameters or a label indicating a characteristic of the critical points. The proposed conditional flow\u2010model allows the generation of points with the desired characteristics. This is of extreme importance in the analysis of equilibrium states and possible transitions, controlled or uncontrolled, to other equilibrium states. Some illustrative examples of functions with hundreds of local minima are used to test the potentialities of the proposed approach. It will be shown that the use of a generative approach is advantageous to explore more complex landscapes compared to a basic random local search algorithm. When applied to the analysis of the uncertain five body problem, the proposed generative model is shown to successfully identify all dynamical equilibrium solutions under uncertainty. Finally when trained on the dynamical stability properties of the critical points, the model can successfully differentiate between stable and metastable solutions. These results show that, for certain types of system, the flow\u2010based model can be trained to find equilibrium points more efficiently than a simple random search. Moreover, we demonstrate that conditional flow\u2010based models are capable of one\u2010shot sampling for specific values of uncertain parameters or characteristics of the equilibrium points.<\/jats:p>","DOI":"10.1111\/exsy.70196","type":"journal-article","created":{"date-parts":[[2026,1,29]],"date-time":"2026-01-29T07:36:11Z","timestamp":1769672171000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Generating Stable and Metastable Critical Points in Uncertain Systems via Flow\u2010Based Models"],"prefix":"10.1111","volume":"43","author":[{"given":"Callum","family":"Wilson","sequence":"first","affiliation":[{"name":"Department of Mechanical and Aerospace Engineering University of Strathclyde  Glasgow UK"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8302-6465","authenticated-orcid":false,"given":"Massimiliano","family":"Vasile","sequence":"additional","affiliation":[{"name":"Department of Mechanical and Aerospace Engineering University of Strathclyde  Glasgow UK"}]}],"member":"311","published-online":{"date-parts":[[2026,1,28]]},"reference":[{"key":"e_1_2_9_2_1","unstructured":"Atanov A. A.Volokhova A.Ashukha I.Sosnovik andD.Vetrov.2020.\u201cSemi\u2010Conditional Normalizing Flows for Semi\u2010Supervised Learning.\u201darXiv.http:\/\/arxiv.org\/abs\/1905.00505arXiv:1905.00505 [cs stat].https:\/\/doi.org\/10.48550\/arXiv.1905.00505."},{"key":"e_1_2_9_3_1","unstructured":"Beeson R. A.Li andA.Sinha.2024.\u201cGlobal Search of Optimal Spacecraft Trajectories Using Amortization and Deep Generative Models.\u201darXiv.http:\/\/arxiv.org\/abs\/2412.20023arXiv:2412.20023 [math].https:\/\/doi.org\/10.48550\/arXiv.2412.20023."},{"key":"e_1_2_9_4_1","doi-asserted-by":"publisher","DOI":"10.1002\/qua.24890"},{"key":"e_1_2_9_5_1","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.98.146401"},{"key":"e_1_2_9_6_1","unstructured":"Bonasera S.2022.Incorporating machine learning into trajectory design strategies in multi\u2010body systems (Unpublished doctoral dissertation). University of Colorado at Boulder."},{"key":"e_1_2_9_7_1","volume-title":"Advances in Neural Information Processing Systems","author":"Chen R. T. Q.","year":"2018"},{"key":"e_1_2_9_8_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10915\u2010022\u201001939\u2010z"},{"key":"e_1_2_9_9_1","unstructured":"Dinh L. D.Krueger andY.Bengio.2015.\u201cNICE: Non\u2010Linear Independent Components Estimation.\u201darXiv.http:\/\/arxiv.org\/abs\/1410.8516arXiv:1410.8516 [cs].https:\/\/doi.org\/10.48550\/arXiv.1410.8516."},{"key":"e_1_2_9_10_1","unstructured":"Dinh L. J.Sohl\u2010Dickstein andS.Bengio.2017.\u201cDensity Estimation Using Real NVP.\u201darXiv.http:\/\/arxiv.org\/abs\/1605.08803arXiv:1605.08803 [cs stat].https:\/\/doi.org\/10.48550\/arXiv.1605.08803."},{"key":"e_1_2_9_11_1","doi-asserted-by":"publisher","DOI":"10.1098\/rsta.2021.0213"},{"key":"e_1_2_9_12_1","volume-title":"Advances in Neural Information Processing Systems","author":"Goodfellow I.","year":"2014"},{"key":"e_1_2_9_13_1","doi-asserted-by":"crossref","unstructured":"Graebner J. andR.Beeson.2025.\u201cGlobal Search for Optimal Low Thrust Spacecraft Trajectories Using Diffusion Models and the Indirect Method.\u201darXiv.http:\/\/arxiv.org\/abs\/2501.07005arXiv:2501.07005 [eess].https:\/\/doi.org\/10.48550\/arXiv.2501.07005.","DOI":"10.1007\/s40295-025-00535-1"},{"key":"e_1_2_9_14_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.swevo.2011.08.003"},{"key":"e_1_2_9_15_1","first-page":"6840","volume-title":"Advances in Neural Information Processing Systems","author":"Ho J.","year":"2020"},{"key":"e_1_2_9_16_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.commatsci.2020.110187"},{"key":"e_1_2_9_17_1","first-page":"4615","volume-title":"Proceedings of the 37th International Conference on Machine Learning","author":"Izmailov P.","year":"2020"},{"key":"e_1_2_9_18_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.asoc.2017.03.042"},{"key":"e_1_2_9_19_1","volume-title":"Advances in Neural Information Processing Systems","author":"Kingma D. P.","year":"2018"},{"key":"e_1_2_9_20_1","unstructured":"Kingma D. P. andM.Welling.2022.\u201cAuto\u2010Encoding Variational Bayes.\u201darXiv.http:\/\/arxiv.org\/abs\/1312.6114arXiv:1312.6114 [cs stat].https:\/\/doi.org\/10.48550\/arXiv.1312.6114."},{"key":"e_1_2_9_21_1","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2020.2992934"},{"key":"e_1_2_9_22_1","volume-title":"Estimation of Distribution Algorithms: A New Tool for Evolutionary Computation","author":"Larra\u00f1aga P.","year":"2001"},{"key":"e_1_2_9_23_1","doi-asserted-by":"publisher","DOI":"10.1109\/CEC.2010.5586349"},{"key":"e_1_2_9_24_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.asr.2015.04.017"},{"key":"e_1_2_9_25_1","doi-asserted-by":"publisher","DOI":"10.1017\/jfm.2023.720"},{"key":"e_1_2_9_26_1","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevE.104.025205"},{"key":"e_1_2_9_27_1","unstructured":"Mirza M. andS.Osindero.2014.\u201cConditional Generative Adversarial Nets.\u201darXiv.http:\/\/arxiv.org\/abs\/1411.1784arXiv:1411.1784 [cs stat].https:\/\/doi.org\/10.48550\/arXiv.1411.1784."},{"key":"e_1_2_9_28_1","first-page":"4723","volume-title":"Proceedings of the 36th International Conference on Machine Learning","author":"Nalisnick E.","year":"2019"},{"key":"e_1_2_9_29_1","doi-asserted-by":"publisher","DOI":"10.1126\/science.aaw1147"},{"key":"e_1_2_9_30_1","doi-asserted-by":"publisher","DOI":"10.1016\/S0747\u20107171(88)80027\u20100"},{"issue":"1","key":"e_1_2_9_31_1","first-page":"2617","article-title":"Normalizing Flows for Probabilistic Modeling and Inference","volume":"22","author":"Papamakarios G.","year":"2021","journal-title":"Journal of Machine Learning Research"},{"key":"e_1_2_9_32_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41467\u2010019\u201010343\u20105"},{"key":"e_1_2_9_33_1","unstructured":"Raveri M. C.Doux andS.Pandey.2024.\u201cUnderstanding Posterior Projection Effects With Normalizing Flows.\u201darXiv.http:\/\/arxiv.org\/abs\/2409.09101.https:\/\/doi.org\/10.48550\/arXiv.2409.09101."},{"key":"e_1_2_9_34_1","first-page":"1530","volume-title":"Proceedings of the 32nd International Conference on Machine Learning","author":"Rezende D.","year":"2015"},{"key":"e_1_2_9_35_1","unstructured":"Rojas C. J. G. A.Dengel andM. D.Ribeiro.2021.\u201cReduced\u2010Order Model for Fluid Flows via Neural Ordinary Differential Equations.\u201darXiv.http:\/\/arxiv.org\/abs\/2102.02248arXiv:2102.02248 [physics].https:\/\/doi.org\/10.48550\/arXiv.2102.02248."},{"key":"e_1_2_9_36_1","volume-title":"Advances in Neural Information Processing Systems","author":"Rubanova Y.","year":"2019"},{"key":"e_1_2_9_37_1","first-page":"30523","volume-title":"Proceedings of the 40th International Conference on Machine Learning","author":"Seifner P.","year":"2023"},{"key":"e_1_2_9_38_1","first-page":"4263","article-title":"Parallel Sampling of Diffusion Models","volume":"36","author":"Shih A.","year":"2023","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_2_9_39_1","unstructured":"Springenberg J. T.2016.\u201cUnsupervised and Semisupervised Learning With Categorical Generative Adversarial Networks.\u201darXiv.http:\/\/arxiv.org\/abs\/1511.06390arXiv:1511.06390 [cs stat].https:\/\/doi.org\/10.48550\/arXiv.1511.06390."},{"key":"e_1_2_9_40_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.asr.2023.01.036"},{"key":"e_1_2_9_41_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10994\u2010019\u201005855\u20106"},{"key":"e_1_2_9_42_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41592\u2010019\u20100686\u20102"},{"key":"e_1_2_9_43_1","volume-title":"Advances in Neural Information Processing Systems","author":"Watter M.","year":"2015"},{"key":"e_1_2_9_44_1","doi-asserted-by":"publisher","DOI":"10.1016\/0004\u20103702(95)00124\u20107"},{"key":"e_1_2_9_45_1","doi-asserted-by":"publisher","DOI":"10.52202\/078368-0083"},{"key":"e_1_2_9_46_1","doi-asserted-by":"publisher","DOI":"10.1038\/s42256\u2010024\u201000837\u20103"},{"key":"e_1_2_9_47_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcp.2019.05.024"},{"key":"e_1_2_9_48_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10509-017-3240-7"}],"container-title":["Expert Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1111\/exsy.70196","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1111\/exsy.70196","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1111\/exsy.70196","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,8]],"date-time":"2026-03-08T23:18:52Z","timestamp":1773011932000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1111\/exsy.70196"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1,28]]},"references-count":47,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2026,3]]}},"alternative-id":["10.1111\/exsy.70196"],"URL":"https:\/\/doi.org\/10.1111\/exsy.70196","archive":["Portico"],"relation":{},"ISSN":["0266-4720","1468-0394"],"issn-type":[{"value":"0266-4720","type":"print"},{"value":"1468-0394","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,1,28]]},"assertion":[{"value":"2025-05-26","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-12-24","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-01-28","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e70196"}}