{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,29]],"date-time":"2026-07-29T07:59:15Z","timestamp":1785311955227,"version":"3.55.0"},"reference-count":48,"publisher":"Society for Industrial & Applied Mathematics (SIAM)","issue":"4","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["SIAM J. Sci. Comput."],"published-print":{"date-parts":[[2026,8,31]]},"abstract":"<jats:p>Abstract.<\/jats:p>\n                  <jats:p>Projection-based model reduction is among the most widely adopted methods for constructing parametric reduced-order models (ROMs).\u00a0Utilizing the snapshot data from solving full-order governing equations, the proper orthogonal decomposition (POD) computes the optimal basis modes that represent the data, and a ROM can be constructed in the low-dimensional vector subspace spanned by the POD basis. For parametric governing equations, a potential challenge arises when there is a need to update the POD basis to adapt ROMs\u00a0that accurately capture the variation of a system\u2019s behavior over its parameter space (in design, control, uncertainty quantification, digital twins applications, etc.). In this paper, we propose a projected Gaussian process (pGP) and formulate the problem of adapting the POD basis as a supervised statistical learning problem, for which the goal is to learn a mapping from the parameter space to the Grassmann manifold that contains the optimal subspaces.\u00a0A mapping is first established between the Euclidean space and the horizontal space of an orthogonal matrix that spans a reference subspace in the Grassmann manifold.\u00a0A second mapping from the horizontal space to the Grassmann manifold is established through the exponential\/logarithm maps between the manifold and its tangent space. Finally, given a new parameter, the conditional distribution of a vector can be found in the Euclidean space using the Gaussian process (GP) regression, and such a distribution is then projected to the Grassmann manifold that enables us to predict the optimal subspace for the new parameter.\u00a0As a statistical learning approach, the proposed pGP allows us to optimally estimate (or tune) the model parameters from data and quantify the statistical uncertainty associated with the prediction. The advantages of the proposed pGP are demonstrated by numerical experiments.<\/jats:p>\n                  <jats:p>Reproducibility of computational results. This paper has been awarded the \u201cSIAM Reproducibility Badge:\u00a0Code and data available\u201d as a recognition that the authors have followed reproducibility principles valued by SISC and the scientific computing community. Code and data that allow readers to reproduce the results in this paper are available at https:\/\/github.com\/dnncode\/pGP and in the supplementary material ( pGP-main.zip [6.24KB]). [Formula: see text]<\/jats:p>","DOI":"10.1137\/24m1697827","type":"journal-article","created":{"date-parts":[[2026,7,29]],"date-time":"2026-07-29T07:01:06Z","timestamp":1785308466000},"page":"C941-C966","source":"Crossref","is-referenced-by-count":0,"title":["Adapting Projection-Based Reduced-Order Models Using Projected Gaussian Process"],"prefix":"10.1137","volume":"48","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-4510-2847","authenticated-orcid":true,"given":"Xiao","family":"Liu","sequence":"first","affiliation":[{"name":"H. Milton Stewart School of Industrial and Systems Engineering, Georgia Tech, Atlanta, GA 30332 USA."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jingyi","family":"Feng","sequence":"additional","affiliation":[{"name":"H. Milton Stewart School of Industrial and Systems Engineering, Georgia Tech, Atlanta, GA 30332 USA."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xinchao","family":"Liu","sequence":"additional","affiliation":[{"name":"H. Milton Stewart School of Industrial and Systems Engineering, Georgia Tech, Atlanta, GA 30332 USA."}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"351","published-online":{"date-parts":[[2026,7,29]]},"reference":[{"key":"ref1","doi-asserted-by":"crossref","unstructured":"D. Amsallem, J. Cortial, and C. Farhat, On-demand CFD-based aeroelastic predictions using a database of reduced-order bases and models, in 47th AIAA Aerospace Sciences Meeting Including the New Horizons Forum and Aerospace Exposition, Orlando, FL, 2009, https:\/\/doi.org\/10.2514\/6.2009-800.","DOI":"10.2514\/6.2009-800"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.2514\/1.35374"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1016\/S0927-0507(06)13018-2"},{"key":"ref4","first-page":"3281","volume":"68","author":"Batool S.","year":"2021","journal-title":"IEEE Trans. Circuits Syst. II Express Briefs"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1137\/090776925"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1007\/s10444-023-10090-8"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-02090-7_6"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1137\/130932715"},{"key":"ref9","doi-asserted-by":"crossref","unstructured":"P. Benner, M. Ohlberger, A. Cohen, and K. Willcox, Model Reduction and Approximation:\u00a0Theory and Algorithms, SIAM, 2017, https:\/\/doi.org\/10.1137\/1.9781611974829.","DOI":"10.1137\/1.9781611974829"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1146\/annurev.fl.25.010193.002543"},{"key":"ref11","unstructured":"J. Berman and B. Peherstorfer, CoLoRA:\u00a0Continuous Low-Rank Adaptation for Reduced Implicit Neural Modeling of Parameterized Partial Differential Equations, preprint, arXiv:2402.14646, 2024."},{"key":"ref12","unstructured":"T. Blau, E. V. Bonilla, I. Chades, and A. Dezfouli, Optimizing sequential experimental design with deep reinforcement learning, in the 39th International Conference on Machine Learning, PMLR, 2022."},{"key":"ref13","doi-asserted-by":"crossref","unstructured":"A. H. J. Borggaard and D. Pelletier, On the use of sensitivity analysis to improve reduced-order models, in 4th Flow Control Conference, Seattle, WA, 2008, https:\/\/doi.org\/10.2514\/6.2008-4192.","DOI":"10.2514\/6.2008-4192"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1016\/j.automatica.2022.110368"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1137\/070694855"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1137\/130916138"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1007\/BF02288367"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1007\/s10444-026-10295-7"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1017\/S0962492902000120"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1007\/s00466-022-02163-0"},{"key":"ref21","doi-asserted-by":"crossref","unstructured":"M. Geuss, H. Panzer, and B. Lohmann, On parametric model order reduction by matrix interpolation, in 2013 European Control Conference (ECC), 2013, pp. 3433\u20133438, https:\/\/doi.org\/10.23919\/ECC.2013.6669829.","DOI":"10.23919\/ECC.2013.6669829"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1016\/j.cma.2020.113269"},{"key":"ref23","doi-asserted-by":"crossref","unstructured":"I. V. Gosea, S. Gugercin, and B. Unger, Parametric model reduction via rational interpolation along parameters, in 2021 60th IEEE Conference on Decision and Control (CDC), 2021, pp. 6895\u20136900, https:\/\/doi.org\/10.1109\/CDC45484.2021.9682841.","DOI":"10.1109\/CDC45484.2021.9682841"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-22470-1"},{"key":"ref25","unstructured":"D. R. Ivanova, A. Foster, S. Kleinegesse, M. U. Gutmann, and T. Rainforth, Implicit deep adaptive design:\u00a0Policy-based experimental design without likelihoods, NeurIPS 2021, 2021."},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1002\/nme.6423"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1137\/1.9781611974508"},{"key":"ref28","doi-asserted-by":"crossref","unstructured":"S. Liao, T. Xue, J. Jeong, S. Webster, K. Ehmann, and J. Cao, Hybrid Full-Field Thermal Characterization of Additive Manufacturing Processes Using Physics-Informed Neural Networks with Data, preprint, arXiv:2206.07756, 2022.","DOI":"10.1007\/s00466-022-02257-9"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.2514\/1.J062180"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1080\/01621459.2017.1409123"},{"key":"ref31","doi-asserted-by":"crossref","unstructured":"A. Mallasto and A. Feragen, Wrapped Gaussian process regression on Riemannian manifolds, in 2018 IEEE\/CVF Conference on Computer Vision and Pattern Recognition, 2018, pp. 5580\u20135588, https:\/\/doi.org\/10.1109\/CVPR.2018.00585.","DOI":"10.1109\/CVPR.2018.00585"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.3934\/dcdss.2019115"},{"key":"ref33","series-title":"Lecture Notes Statist. 118","volume-title":"Bayesian Learning for Neural Networks","author":"Neal R. M.","year":"2012"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1524\/auto.2010.0863"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1016\/j.procs.2015.05.363"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1137\/140989169"},{"key":"ref37","doi-asserted-by":"crossref","unstructured":"S. Prajna, POD model reduction with stability guarantee, in 42nd IEEE International Conference on Decision and Control, Vol. 5, 2003, pp. 5254\u20135258, https:\/\/doi.org\/10.1109\/CDC.2003.1272472.","DOI":"10.1109\/CDC.2003.1272472"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1016\/j.physd.2020.132401"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.7551\/mitpress\/3206.001.0001"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1142\/S0218127405012429"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevE.92.033304"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1002\/nme.4408"},{"key":"ref43","doi-asserted-by":"crossref","unstructured":"D. Sorensen, Passivity preserving model reduction via interpolation of spectral zeros, in 2003 European Control Conference (ECC), 2003, pp. 974\u2013978, https:\/\/doi.org\/10.23919\/ECC.2003.7085085.","DOI":"10.23919\/ECC.2003.7085085"},{"key":"ref44","first-page":"85","volume":"2","author":"Vetrano F.","year":"2011","journal-title":"J. Aeroelasticity Structural Dynam."},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/TAC.2016.2521361"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1016\/j.optlastec.2018.04.034"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1137\/15M1054201"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1137\/21M1432739"}],"container-title":["SIAM Journal on Scientific Computing"],"original-title":[],"language":"en","deposited":{"date-parts":[[2026,7,29]],"date-time":"2026-07-29T07:01:10Z","timestamp":1785308470000},"score":1,"resource":{"primary":{"URL":"https:\/\/epubs.siam.org\/doi\/10.1137\/24M1697827"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,29]]},"references-count":48,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2026,8,31]]}},"alternative-id":["10.1137\/24M1697827"],"URL":"https:\/\/doi.org\/10.1137\/24m1697827","relation":{},"ISSN":["1064-8275","1095-7197"],"issn-type":[{"value":"1064-8275","type":"print"},{"value":"1095-7197","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,29]]}}}