{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T18:24:58Z","timestamp":1787336698306,"version":"build-2736575974"},"reference-count":67,"publisher":"Society for Industrial & Applied Mathematics (SIAM)","issue":"2","funder":[{"name":"National Science Foundation","award":["2012250"],"award-info":[{"award-number":["2012250"]}]},{"name":"National Science Foundation","award":["1901091"],"award-info":[{"award-number":["1901091"]}]},{"name":"National Science Foundation","award":["DMS-1439786"],"award-info":[{"award-number":["DMS-1439786"]}]},{"DOI":"10.13039\/100000893","name":"Simons Foundation","doi-asserted-by":"crossref","award":["50736"],"award-info":[{"award-number":["50736"]}],"id":[{"id":"10.13039\/100000893","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["SIAM J. Sci. Comput."],"published-print":{"date-parts":[[2023,4,30]]},"abstract":"<jats:p>Abstract.<\/jats:p>\n                  <jats:p>Robust controllers that stabilize dynamical systems even under disturbances and noise are often formulated as solutions of nonsmooth, nonconvex optimization problems. While methods such as gradient sampling can handle the nonconvexity and nonsmoothness, the costs of evaluating the objective function may be substantial, making robust control challenging for dynamical systems with high-dimensional state spaces. In this work, we introduce multifidelity variants of gradient sampling that leverage low-cost, low-fidelity models with low-dimensional state spaces for speeding up the optimization process while nonetheless providing convergence guarantees for a high-fidelity model of the system of interest, which is primarily accessed in the last phase of the optimization process. Our first multifidelity method initiates gradient sampling on higher-fidelity models with starting points obtained from cheaper, lower-fidelity models. Our second multifidelity method relies on ensembles of gradients that are computed from low- and high-fidelity models. Numerical experiments with controlling the cooling of a steel rail profile and laminar flow in a cylinder wake demonstrate that our new multifidelity gradient sampling methods achieve up to two orders of magnitude speedup compared to the single-fidelity gradient sampling method that relies on the high-fidelity model alone.<\/jats:p>","DOI":"10.1137\/22m1500137","type":"journal-article","created":{"date-parts":[[2023,4,28]],"date-time":"2023-04-28T05:51:33Z","timestamp":1682661093000},"page":"A933-A957","source":"Crossref","is-referenced-by-count":2,"title":["Multifidelity Robust Controller Design with Gradient Sampling"],"prefix":"10.1137","volume":"45","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-1667-4862","authenticated-orcid":true,"given":"Steffen W. R.","family":"Werner","sequence":"first","affiliation":[{"name":"Courant Institute of Mathematical Sciences, New York University, New York, NY 10012 USA."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6563-6371","authenticated-orcid":true,"given":"Michael L.","family":"Overton","sequence":"additional","affiliation":[{"name":"Courant Institute of Mathematical Sciences, New York University, New York, NY 10012 USA."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1558-6775","authenticated-orcid":true,"given":"Benjamin","family":"Peherstorfer","sequence":"additional","affiliation":[{"name":"Courant Institute of Mathematical Sciences, New York University, New York, NY 10012 USA."}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"351","published-online":{"date-parts":[[2023,4,28]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1007\/BF01197433"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1137\/19M125892X"},{"key":"ref3","unstructured":"T. Alsup , \nL. Venturi , and \nB. Peherstorfer , Multilevel Stein variational gradient descent with applications to Bayesian inverse problems, in Proceedings of the 2nd Mathematical and Scientific Machine Learning Conference, Proc. Mach. Learn. Res. (PMLR) 145, \nJ. Bruna , \nJ. Hesthaven , and \nL. Zdeborova , eds. JMLR, Cambridge, MA, 2022, pp. 93\u2013117, https:\/\/proceedings.mlr.press\/v145\/alsup22a.html."},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1137\/1.9780898718713"},{"key":"ref5","doi-asserted-by":"crossref","unstructured":"E. Arian , \nM. Fahl , and \nE. W. Sachs , Managing POD models by optimization methods, in Proceedings of the 41st IEEE Conference on Decision and Control, 2002, pp. 3300\u20133305, https:\/\/doi.org\/10.1109\/CDC.2002.1184383.","DOI":"10.1109\/CDC.2002.1184383"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-72040-7_2"},{"key":"ref7","volume-title":"Example Setups of Navier-Stokes Equations With Control and Observation: Spatial Discretization and Representation via Linear-Quadratic Matrix Coefficients","author":"Behr M.","year":"2017"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/MCS.2004.1272745"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1137\/1.9781611974829"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1137\/130932715"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1007\/s11075-022-01331-w"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1007\/s10589-022-00359-x"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1002\/nla.622"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1137\/17M1137966"},{"key":"ref15","doi-asserted-by":"crossref","unstructured":"P. Benner , \nT. Mitchell , and \nM. L. Overton , Low-order control design using a reduced-order model with a stability constraint on the full-order model, in Proceedings of the 2018 IEEE Conference on Decision and Control (CDC), 2018, pp. 3000\u20133005, https:\/\/doi.org\/10.1109\/CDC.2018.8619449.","DOI":"10.1109\/CDC.2018.8619449"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1515\/9783110498967"},{"key":"ref17","volume-title":"Model Order Reduction. Volume 2: Snapshot-Based Methods and Algorithms","author":"Benner P.","year":"2021"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1007\/978-0-387-31256-9"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1007\/s10589-005-3228-z"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-34910-3_6"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.3182\/20060705-3-FR-2907.00059"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/TAC.2006.884944"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1016\/S0024-3795(02)00260-4"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1016\/S1474-6670(17)35659-8"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1137\/030601296"},{"key":"ref26","doi-asserted-by":"crossref","unstructured":"A. Chaudhuri , \nB. Peherstorfer , and \nK. Willcox , Multifidelity cross-entropy estimation of conditional value-at-risk for risk-averse design optimization, in Proceedings of the AIAA Scitech 2020 Forum, 2020, AIAA 2020-2129, https:\/\/doi.org\/10.2514\/6.2020-2129.","DOI":"10.2514\/6.2020-2129"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1007\/s00791-011-0160-x"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1080\/10556788.2016.1208749"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/9.29425"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-55508-4_16"},{"key":"ref31","doi-asserted-by":"crossref","unstructured":"C. C. Fischer , \nR. V. Grandhi , and \nP. S. Beran , Bayesian low-fidelity correction approach to multi-fidelity aerospace design, in Proceedings of the 58th AIAA\/ASCE\/AHS\/ASC Structures, Structural Dynamics, and Materials Conference, 2017, AIAA 2017-0132, https:\/\/doi.org\/10.2514\/6.2017-0133.","DOI":"10.2514\/6.2017-0133"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1007\/BFb0007371"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1287\/opre.1070.0496"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.56021\/9781421407944"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1137\/090779127"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1137\/050639673"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1137\/16M1088958"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1088\/1741-4326\/ac4777"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1007\/s10107-012-0514-2"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1007\/s00158-011-0749-1"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.2514\/1.J051125"},{"key":"ref42","unstructured":"M. Millstone , \nM. L. Overton , \nD. Henrion , \nG. Deaconu , \nS. Gumussoy , and \nD. Arzelier , HIFOO: A MATLAB Package for Fixed-order   \\(\\mathcal{H}_{\\infty}\\)   and   \\(\\mathcal{H}_{2}\\)   Controller Design (Version 3.5), 2011, https:\/\/cs.nyu.edu\/overton\/software\/hifoo\/."},{"key":"ref43","unstructured":"T. Mitchell , ROSTAPACK: RObust STAbility PACKage (Version 3.0), 2022, http:\/\/www.timmitchell.com\/software\/ROSTAPACK\/."},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1016\/j.ifacol.2015.09.428"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/9.86941"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1002\/nme.4761"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.2514\/1.C033352"},{"key":"ref48","volume-title":"Communications and Control Engineering","author":"Obinata G.","year":"2001"},{"key":"ref49","unstructured":"M. L. Overton , HANSO: Hybrid Algorithm for Non-Smooth Optimization (Version 3.0), 2021, https:\/\/cs.nyu.edu\/overton\/software\/hanso\/."},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1007\/s10092-021-00448-z"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1002\/nla.1863"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1007\/s00211-018-0945-7"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1016\/j.cma.2015.03.018"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1137\/140989169"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1137\/15M1046472"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1137\/16M1082469"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1137\/16M1081981"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-02090-7"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-540-78841-6_7"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.2514\/1.36043"},{"key":"ref61","volume-title":"Efficient Numerical Solution of Large Scale Algebraic Matrix Equations in PDE Control and Model Order Reduction","author":"Saak J.","year":"2009"},{"key":"ref62","unstructured":"J. Saak , \nM. K\u00f6hler , and \nP. Benner , M-M.E.S.S. \u2013 The Matrix Equations Sparse Solver Library, version 2.1, 2021, https:\/\/doi.org\/10.5281\/zenodo.4719688."},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1137\/S0895479800371529"},{"key":"ref64","unstructured":"S. W. R. Werner , Code, Data and Results for Numerical Experiments in \u201cMultifidelity Robust Controller Design with Gradient Sampling\u201d, version 1.0, 2022, https:\/\/doi.org\/10.5281\/zenodo.6403121."},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.1137\/09074927X"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.1002\/nme.4770"},{"key":"ref67","volume-title":"Essentials of Robust Control","author":"Zhou K.","year":"1998"}],"container-title":["SIAM Journal on Scientific Computing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/epubs.siam.org\/doi\/pdf\/10.1137\/22M1500137","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T17:26:19Z","timestamp":1787333179000},"score":1,"resource":{"primary":{"URL":"https:\/\/epubs.siam.org\/doi\/10.1137\/22M1500137"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,4,28]]},"references-count":67,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2023,4,30]]}},"alternative-id":["10.1137\/22M1500137"],"URL":"https:\/\/doi.org\/10.1137\/22m1500137","relation":{},"ISSN":["1064-8275","1095-7197"],"issn-type":[{"value":"1064-8275","type":"print"},{"value":"1095-7197","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,4,28]]}}}