{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,8]],"date-time":"2026-07-08T15:47:00Z","timestamp":1783525620353,"version":"3.55.0"},"reference-count":46,"publisher":"Springer Science and Business Media LLC","issue":"2","license":[{"start":{"date-parts":[[2025,1,11]],"date-time":"2025-01-11T00:00:00Z","timestamp":1736553600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,1,11]],"date-time":"2025-01-11T00:00:00Z","timestamp":1736553600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100002954","name":"Universit\u00e0 degli Studi di Milano - Bicocca","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100002954","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["J Glob Optim"],"published-print":{"date-parts":[[2025,6]]},"abstract":"<jats:title>Abstract<\/jats:title>\n          <jats:p>Gaussian Process regression is a kernel method successfully adopted in many real-life applications. Recently, there is a growing interest on extending this method to non-Euclidean input spaces, like the one considered in this paper, consisting of probability measures. Although a Positive Definite kernel can be defined by using a suitable distance\u2014the Wasserstein distance\u2014 the common procedure for learning the Gaussian Process model can fail due to numerical issues, arising earlier and more frequently than in the case of an Euclidean input space and, as demonstrated, impossible to avoid by adding artificial noise (<jats:italic>nugget effect<\/jats:italic>) as usually done. This paper uncovers the main reason of these issues, that is a non-stationarity relation between the Wasserstein-based squared exponential kernel and its Euclidean counterpart. As a relevant result, we learn a Gaussian Process model by assuming the input space as Euclidean and then use an algebraic transformation, based on the uncovered relation, to transform it into a non-stationary and Wasserstein-based Gaussian Process model over probability measures. This algebraic transformation is simpler than <jats:italic>log-exp<\/jats:italic> maps used on data belonging to Riemannian manifolds and recently extended to consider the pseudo-Riemannian structure of an input space equipped with the Wasserstein distance.<\/jats:p>","DOI":"10.1007\/s10898-025-01463-y","type":"journal-article","created":{"date-parts":[[2025,1,11]],"date-time":"2025-01-11T05:53:27Z","timestamp":1736574807000},"page":"253-278","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":3,"title":["Gaussian Process regression over discrete probability measures: on the non-stationarity relation between Euclidean and Wasserstein Squared Exponential Kernels"],"prefix":"10.1007","volume":"92","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-1431-576X","authenticated-orcid":false,"given":"Antonio","family":"Candelieri","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Andrea","family":"Ponti","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Francesco","family":"Archetti","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2025,1,11]]},"reference":[{"key":"1463_CR1","unstructured":"Williams, C.K.: Gaussian Processes for Machine Learning, 2, (2006)"},{"key":"1463_CR2","doi-asserted-by":"publisher","DOI":"10.1201\/9780367815493","volume-title":"Surrogates: Gaussian Process Modeling, Design, and Optimization for the Applied Sciences","author":"RB Gramacy","year":"2020","unstructured":"Gramacy, R.B.: Surrogates: Gaussian Process Modeling, Design, and Optimization for the Applied Sciences. Chapman and Hall\/CRC, USA (2020)"},{"key":"1463_CR3","volume-title":"Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond","author":"B Scholkopf","year":"2018","unstructured":"Scholkopf, B., Smola, A.J.: Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond. MIT press, USA (2018)"},{"key":"1463_CR4","doi-asserted-by":"crossref","unstructured":"Frazier, P.I.: Bayesian optimization. In: Recent Advances in Optimization and Modeling of Contemporary Problems, 255\u2013278. Informs, USA (2018)","DOI":"10.1287\/educ.2018.0188"},{"key":"1463_CR5","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-24494-1","volume-title":"Bayesian Optimization and Data Science","author":"F Archetti","year":"2019","unstructured":"Archetti, F., Candelieri, A.: Bayesian Optimization and Data Science. Springer, Germany (2019)"},{"key":"1463_CR6","doi-asserted-by":"crossref","unstructured":"Candelieri, A.: A gentle introduction to bayesian optimization. In: 2021 Winter Simulation Conference (WSC), 1\u201316 (2021). IEEE","DOI":"10.1109\/WSC52266.2021.9715413"},{"key":"1463_CR7","doi-asserted-by":"publisher","DOI":"10.1017\/9781108348973","volume-title":"Bayesian Optimization","author":"R Garnett","year":"2023","unstructured":"Garnett, R.: Bayesian Optimization. Cambridge University Press, UK (2023)"},{"issue":"2","key":"1463_CR8","doi-asserted-by":"publisher","first-page":"408","DOI":"10.1109\/TPAMI.2013.218","volume":"37","author":"MP Deisenroth","year":"2013","unstructured":"Deisenroth, M.P., Fox, D., Rasmussen, C.E.: Gaussian processes for data-efficient learning in robotics and control. IEEE Trans. Pattern Anal. Mach. Intell. 37(2), 408\u2013423 (2013)","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"1463_CR9","unstructured":"Sui, Y., Gotovos, A., Burdick, J., Krause, A.: Safe exploration for optimization with gaussian processes. In: International Conference on Machine Learning, 997\u20131005 (2015). PMLR"},{"key":"1463_CR10","doi-asserted-by":"crossref","unstructured":"Berkenkamp, F., Schoellig, A.P., Krause, A.: Safe controller optimization for quadrotors with gaussian processes. In: 2016 IEEE International Conference on Robotics and Automation (ICRA), 491\u2013496 (2016). IEEE","DOI":"10.1109\/ICRA.2016.7487170"},{"key":"1463_CR11","unstructured":"Jaquier, N., Rozo, L., Calinon, S., B\u00fcrger, M.: Bayesian optimization meets riemannian manifolds in robot learning. In: Conference on Robot Learning, 233\u2013246 (2020). PMLR"},{"key":"1463_CR12","doi-asserted-by":"crossref","unstructured":"Calandra, R., Peters, J., Rasmussen, C.E., Deisenroth, M.P.: Manifold gaussian processes for regression. In: 2016 International Joint Conference on Neural Networks (IJCNN), 3338\u20133345 (2016). IEEE","DOI":"10.1109\/IJCNN.2016.7727626"},{"key":"1463_CR13","doi-asserted-by":"crossref","unstructured":"Mallasto, A., Feragen, A.: Wrapped gaussian process regression on Riemannian manifolds. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2018)","DOI":"10.1109\/CVPR.2018.00585"},{"key":"1463_CR14","doi-asserted-by":"crossref","unstructured":"Feragen, A., Lauze, F., Hauberg, S.: Geodesic exponential kernels: When curvature and linearity conflict. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 3032\u20133042 (2015)","DOI":"10.1109\/CVPR.2015.7298922"},{"key":"1463_CR15","unstructured":"Feragen, A., Hauberg, S.: Open problem: Kernel methods on manifolds and metric spaces. what is the probability of a positive definite geodesic exponential kernel? In: Conference on Learning Theory, 1647\u20131650 (2016). PMLR"},{"issue":"5\u20136","key":"1463_CR16","doi-asserted-by":"publisher","first-page":"355","DOI":"10.1561\/2200000073","volume":"11","author":"G Peyr\u00e9","year":"2019","unstructured":"Peyr\u00e9, G., Cuturi, M., et al.: Computational optimal transport: With applications to data science. Found. Trends Mach. Learn. 11(5\u20136), 355\u2013607 (2019)","journal-title":"Found. Trends Mach. Learn."},{"key":"1463_CR17","unstructured":"Villani, C.: Topics in Optimal Transportation, 58. American Mathematical Society, USA (2021)"},{"key":"1463_CR18","doi-asserted-by":"crossref","unstructured":"Simon, D., Aberdam, A.: Barycenters of natural images constrained wasserstein barycenters for image morphing. In: Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, 7910\u20137919 (2020)","DOI":"10.1109\/CVPR42600.2020.00793"},{"key":"1463_CR19","doi-asserted-by":"crossref","unstructured":"Chen, Y., Li, C., Lu, Z.: Computing wasserstein-p distance between images with linear cost. In: Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, 519\u2013528 (2022)","DOI":"10.1109\/CVPR52688.2022.00060"},{"key":"1463_CR20","volume":"10","author":"A Ponti","year":"2021","unstructured":"Ponti, A., Candelieri, A., Archetti, F.: A Wasserstein distance based multiobjective evolutionary algorithm for the risk aware optimization of sensor placement. Intell. Syst. Appl. 10, 200047 (2021)","journal-title":"Intell. Syst. Appl."},{"issue":"4","key":"1463_CR21","doi-asserted-by":"publisher","first-page":"138","DOI":"10.3390\/bdcc6040138","volume":"6","author":"A Ponti","year":"2022","unstructured":"Ponti, A., Giordani, I., Mistri, M., Candelieri, A., Archetti, F.: The unreasonable effectiveness of the Wasserstein distance in analyzing key performance indicators of a network of stores. Big Data Cognit. Comput. 6(4), 138 (2022)","journal-title":"Big Data Cognit. Comput."},{"key":"1463_CR22","unstructured":"Candelieri, A., Ponti, A., Giordani, I., Archetti, F.: On the use of Wasserstein distance in the distributional analysis of human decision making under uncertainty. Annals of Mathematics and Artificial Intelligence, 1\u201322 (2022)"},{"key":"1463_CR23","doi-asserted-by":"crossref","unstructured":"Candelieri, A., Ponti, A., Giordani, I., Bosio, A., Archetti, F.: Distributional learning in multi-objective optimization of recommender systems. J. Ambient Intell. Humanized Comput., 1\u201317 (2022)","DOI":"10.1007\/s12652-022-04356-0"},{"key":"1463_CR24","unstructured":"Kandasamy, K., Neiswanger, W., Schneider, J., Poczos, B., Xing, E.P.: Neural architecture search with bayesian optimisation and optimal transport. Adv. Neural Inf. Process. Syst. 31 (2018)"},{"key":"1463_CR25","unstructured":"Nguyen, V., Le, T., Yamada, M., Osborne, M.A.: Optimal transport kernels for sequential and parallel neural architecture search. In: International Conference on Machine Learning, 8084\u20138095 (2021). PMLR"},{"key":"1463_CR26","doi-asserted-by":"crossref","unstructured":"Bigot, J., Gouet, R., Klein, T., L\u00f3pez, A.: Geodesic pca in the wasserstein space by convex pca. In: Annales de l\u2019Institut Henri Poincar\u00e9, Probabilit\u00e9s et Statistiques, 53, 1\u201326 (2017). Institut Henri Poincar\u00e9","DOI":"10.1214\/15-AIHP706"},{"issue":"2","key":"1463_CR27","doi-asserted-by":"publisher","first-page":"429","DOI":"10.1137\/17M1143459","volume":"40","author":"E Cazelles","year":"2018","unstructured":"Cazelles, E., Seguy, V., Bigot, J., Cuturi, M., Papadakis, N.: Geodesic pca versus log-pca of histograms in the Wasserstein space. SIAM J. Sci. Comput. 40(2), 429\u2013456 (2018)","journal-title":"SIAM J. Sci. Comput."},{"issue":"10","key":"1463_CR28","doi-asserted-by":"publisher","first-page":"6620","DOI":"10.1109\/TIT.2017.2762322","volume":"64","author":"F Bachoc","year":"2017","unstructured":"Bachoc, F., Gamboa, F., Loubes, J.-M., Venet, N.: A gaussian process regression model for distribution inputs. IEEE Trans. Inf. Theory 64(10), 6620\u20136637 (2017)","journal-title":"IEEE Trans. Inf. Theory"},{"key":"1463_CR29","unstructured":"Oh, J.H., Pouryahya, M., Iyer, A., Apte, A.P., Tannenbaum, A., Deasy, J.O.: Kernel Wasserstein distance. arXiv preprint arXiv:1905.09314 (2019)"},{"issue":"7","key":"1463_CR30","doi-asserted-by":"publisher","first-page":"1741","DOI":"10.1109\/TPAMI.2019.2903050","volume":"42","author":"Z Zhang","year":"2019","unstructured":"Zhang, Z., Wang, M., Nehorai, A.: Optimal transport in reproducing kernel Hilbert spaces: theory and applications. IEEE Trans. Pattern Anal. Mach. Intell. 42(7), 1741\u20131754 (2019)","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"1463_CR31","doi-asserted-by":"crossref","unstructured":"De\u00a0Plaen, H., Fanuel, M., Suykens, J.A.: Wasserstein exponential kernels. In: 2020 International Joint Conference on Neural Networks (IJCNN), 1\u20136 (2020). IEEE","DOI":"10.1109\/IJCNN48605.2020.9207630"},{"issue":"1","key":"1463_CR32","first-page":"761","volume":"6","author":"D Higdon","year":"1999","unstructured":"Higdon, D., Swall, J., Kern, J.: Non-stationary spatial modeling. Bayesian Stat. 6(1), 761\u2013768 (1999)","journal-title":"Bayesian Stat."},{"issue":"3","key":"1463_CR33","doi-asserted-by":"publisher","first-page":"743","DOI":"10.1111\/1467-9868.00413","volume":"65","author":"AM Schmidt","year":"2003","unstructured":"Schmidt, A.M., O\u2019Hagan, A.: Bayesian inference for non-stationary spatial covariance structure via spatial deformations. J. R. Stat. Soc.: Ser. B (Statistical Methodology) 65(3), 743\u2013758 (2003)","journal-title":"J. R. Stat. Soc.: Ser. B (Statistical Methodology)"},{"issue":"1","key":"1463_CR34","doi-asserted-by":"publisher","first-page":"321","DOI":"10.1007\/s11081-020-09517-8","volume":"22","author":"A Hebbal","year":"2021","unstructured":"Hebbal, A., Brevault, L., Balesdent, M., Talbi, E.-G., Melab, N.: Bayesian optimization using deep gaussian processes with applications to aerospace system design. Optim. Eng. 22(1), 321\u2013361 (2021)","journal-title":"Optim. Eng."},{"issue":"470","key":"1463_CR35","doi-asserted-by":"publisher","first-page":"653","DOI":"10.1198\/016214504000002014","volume":"100","author":"H-M Kim","year":"2005","unstructured":"Kim, H.-M., Mallick, B.K., Holmes, C.C.: Analyzing nonstationary spatial data using piecewise gaussian processes. J. Am. Stat. Assoc. 100(470), 653\u2013668 (2005)","journal-title":"J. Am. Stat. Assoc."},{"issue":"483","key":"1463_CR36","doi-asserted-by":"publisher","first-page":"1119","DOI":"10.1198\/016214508000000689","volume":"103","author":"RB Gramacy","year":"2008","unstructured":"Gramacy, R.B., Lee, H.K.H.: Bayesian treed gaussian process models with an application to computer modeling. J. Am. Stat. Assoc. 103(483), 1119\u20131130 (2008)","journal-title":"J. Am. Stat. Assoc."},{"key":"1463_CR37","doi-asserted-by":"publisher","first-page":"1","DOI":"10.18637\/jss.v019.i09","volume":"19","author":"RB Gramacy","year":"2007","unstructured":"Gramacy, R.B.: tgp: an r package for Bayesian nonstationary, semiparametric nonlinear regression and design by treed gaussian process models. J. Stat. Softw. 19, 1\u201346 (2007)","journal-title":"J. Stat. Softw."},{"key":"1463_CR38","volume-title":"The Mesh Adaptive Direct Search Algorithm with Treed Gaussian Process Surrogates","author":"RB Gramacy","year":"2011","unstructured":"Gramacy, R.B., Le Digabel, S.: The Mesh Adaptive Direct Search Algorithm with Treed Gaussian Process Surrogates. Groupe d\u2019\u00e9tudes et de recherche en analyse des d\u00e9cisions, France (2011)"},{"key":"1463_CR39","doi-asserted-by":"crossref","unstructured":"Civera, M., Surace, C., Worden, K.: Detection of cracks in beams using treed gaussian processes. In: Structural Health Monitoring & Damage Detection. 7, 85\u201397. Springer, Germany (2017)","DOI":"10.1007\/978-3-319-54109-9_10"},{"key":"1463_CR40","doi-asserted-by":"publisher","DOI":"10.1016\/j.compstruct.2020.112882","volume":"254","author":"M Civera","year":"2020","unstructured":"Civera, M., Boscato, G., Fragonara, L.Z.: Treed gaussian process for manufacturing imperfection identification of pultruded gfrp thin-walled profile. Compos. Struct. 254, 112882 (2020)","journal-title":"Compos. Struct."},{"key":"1463_CR41","doi-asserted-by":"crossref","unstructured":"Candelieri, A., Pedrielli, G.: Treed-gaussian processes with support vector machines as nodes for nonstationary bayesian optimization. In: 2021 Winter Simulation Conference (WSC), 1\u201312 (2021). IEEE","DOI":"10.1109\/WSC52266.2021.9715514"},{"key":"1463_CR42","unstructured":"Dolgov, M., Hanebeck, U.D.: A distance-based framework for gaussian processes over probability distributions. arXiv preprint arXiv:1809.09193 (2018)"},{"issue":"1","key":"1463_CR43","first-page":"533","volume":"15","author":"M Cuturi","year":"2014","unstructured":"Cuturi, M., Avis, D.: Ground metric learning. J. Mach. Learn. Res. 15(1), 533\u2013564 (2014)","journal-title":"J. Mach. Learn. Res."},{"issue":"1","key":"1463_CR44","doi-asserted-by":"publisher","first-page":"89","DOI":"10.1007\/s10851-020-00996-z","volume":"63","author":"M Heitz","year":"2021","unstructured":"Heitz, M., Bonneel, N., Coeurjolly, D., Cuturi, M., Peyr\u00e9, G.: Ground metric learning on graphs. J. Math. Imaging Vision 63(1), 89\u2013107 (2021)","journal-title":"J. Math. Imaging Vision"},{"key":"1463_CR45","unstructured":"Huizing, G.-J., Cantini, L., Peyr\u00e9, G.: Unsupervised ground metric learning using wasserstein singular vectors. In: International Conference on Machine Learning, 9429\u20139443 (2022). PMLR"},{"key":"1463_CR46","unstructured":"Crovini, E., Cotter, S.L., Zygalakis, K., Duncan, A.B.: Batch bayesian optimization via particle gradient flows. arXiv preprint arXiv:2209.04722 (2022)"}],"container-title":["Journal of Global Optimization"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s10898-025-01463-y.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s10898-025-01463-y\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s10898-025-01463-y.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,2]],"date-time":"2025-06-02T06:28:51Z","timestamp":1748845731000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s10898-025-01463-y"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,1,11]]},"references-count":46,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2025,6]]}},"alternative-id":["1463"],"URL":"https:\/\/doi.org\/10.1007\/s10898-025-01463-y","relation":{},"ISSN":["0925-5001","1573-2916"],"issn-type":[{"value":"0925-5001","type":"print"},{"value":"1573-2916","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,1,11]]},"assertion":[{"value":"4 July 2024","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"2 January 2025","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"11 January 2025","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"Authors declare no conflict of interest.","order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}},{"value":"Not applicable.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval"}},{"value":"Not applicable.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent to participate"}},{"value":"Not applicable.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}}]}}