{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,21]],"date-time":"2026-04-21T18:21:47Z","timestamp":1776795707438,"version":"3.51.2"},"reference-count":50,"publisher":"IOP Publishing","issue":"4","license":[{"start":{"date-parts":[[2025,10,27]],"date-time":"2025-10-27T00:00:00Z","timestamp":1761523200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2025,10,27]],"date-time":"2025-10-27T00:00:00Z","timestamp":1761523200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"DOI":"10.13039\/501100001659","name":"Deutsche Forschungsgemeinschaft","doi-asserted-by":"crossref","award":["496900167"],"award-info":[{"award-number":["496900167"]}],"id":[{"id":"10.13039\/501100001659","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Mach. Learn.: Sci. Technol."],"published-print":{"date-parts":[[2025,12,30]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Uncertainty estimations for machine learning interatomic potentials (MLIPs) are crucial for quantifying model error and identifying informative training samples in active learning (AL) strategies. In this study, we evaluate uncertainty estimations of Gaussian process regression (GPR)-based MLIPs, including the predictive GPR standard deviation and ensemble-based uncertainties. We do this in terms of calibration and in terms of impact on model performance in an AL scheme. We consider GPR models with Coulomb and smooth overlap of atomic positions representations as inputs to predict potential energy surfaces and excitation energies of molecules. Regarding calibration, we find that ensemble-based uncertainty estimations show already poor global calibration (e.g. averaged over the whole test set). In contrast, the GPR standard deviation shows good global calibration, but when grouping predictions by their uncertainty, we observe a systematical bias for predictions with high uncertainty. Although an increasing uncertainty correlates with an increasing bias, the bias is not captured quantitatively by the uncertainty. Therefore, the GPR standard deviation can be useful to identify predictions with a high bias and error but, without further knowledge, should not be interpreted as a quantitative measure for a potential error range. Selecting the samples with the highest GPR standard deviation from a fixed configuration space leads to a model that overemphasizes the borders of the configuration space represented in the fixed dataset. This may result in worse performance in more densely sampled areas but better generalization for extrapolation tasks.<\/jats:p>","DOI":"10.1088\/2632-2153\/ae09ef","type":"journal-article","created":{"date-parts":[[2025,9,22]],"date-time":"2025-09-22T22:52:23Z","timestamp":1758581543000},"page":"045019","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Evaluation of uncertainty estimations for Gaussian process regression based machine learning interatomic potentials"],"prefix":"10.1088","volume":"6","author":[{"ORCID":"https:\/\/orcid.org\/0009-0000-9174-1178","authenticated-orcid":true,"given":"Matthias","family":"Holzenkamp","sequence":"first","affiliation":[]},{"given":"Dongyu","family":"Lyu","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6114-7431","authenticated-orcid":true,"given":"Ulrich","family":"Kleinekath\u00f6fer","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7028-6580","authenticated-orcid":true,"given":"Peter","family":"Zaspel","sequence":"additional","affiliation":[]}],"member":"266","published-online":{"date-parts":[[2025,10,27]]},"reference":[{"key":"mlstae09efbib1","doi-asserted-by":"publisher","DOI":"10.1063\/1.5019779","type":"journal-article","volume":"148","author":"Sch\u00fctt","year":"2018","journal-title":"J. Chem. Phys."},{"key":"mlstae09efbib2","first-page":"pp 9377","type":"conference-proceedings","article-title":"Equivariant message passing for the prediction of tensorial properties and molecular spectra","author":"Sch\u00fctt","year":"2021"},{"key":"mlstae09efbib3","doi-asserted-by":"publisher","first-page":"2453","DOI":"10.1038\/s41467-022-29939-5","type":"journal-article","volume":"13","author":"Batzner","year":"2022","journal-title":"Nat. Commun."},{"key":"mlstae09efbib4","first-page":"pp 11423","type":"conference-proceedings","volume":"vol 35","author":"Batatia","year":"2022"},{"key":"mlstae09efbib5","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.108.058301","type":"journal-article","volume":"108","author":"Rupp","year":"2012","journal-title":"Phys. Rev. Lett."},{"key":"mlstae09efbib6","doi-asserted-by":"publisher","DOI":"10.1088\/2632-2153\/aca005","type":"journal-article","volume":"3","author":"Huo","year":"2022","journal-title":"Mach. Learn.: Sci. Technol."},{"key":"mlstae09efbib7","doi-asserted-by":"publisher","first-page":"2326","DOI":"10.1021\/acs.jpclett.5b00831","type":"journal-article","volume":"6","author":"Hansen","year":"2015","journal-title":"J. Phys. Chem. Lett."},{"key":"mlstae09efbib8","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevB.87.184115","type":"journal-article","volume":"87","author":"Bart\u00f3k","year":"2013","journal-title":"Phys. Rev. B"},{"key":"mlstae09efbib9","doi-asserted-by":"publisher","first-page":"316","DOI":"10.1016\/j.jcp.2014.12.018","type":"journal-article","volume":"285","author":"Thompson","year":"2015","journal-title":"J. Comput. Phys."},{"key":"mlstae09efbib10","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevB.99.014104","type":"journal-article","volume":"99","author":"Drautz","year":"2019","journal-title":"Phys. Rev. B"},{"key":"mlstae09efbib11","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.98.146401","type":"journal-article","volume":"98","author":"Behler","year":"2007","journal-title":"Phys. Rev. Lett."},{"key":"mlstae09efbib12","doi-asserted-by":"publisher","first-page":"41","DOI":"10.1038\/s41524-022-00721-x","type":"journal-article","volume":"8","author":"Langer","year":"2022","journal-title":"npj Comput. Mater."},{"key":"mlstae09efbib13","doi-asserted-by":"publisher","first-page":"7696","DOI":"10.1021\/acs.jctc.1c00647","type":"journal-article","volume":"17","author":"Kov\u00e1cs","year":"2021","journal-title":"J. Chem. Theory Comput."},{"key":"mlstae09efbib14","doi-asserted-by":"publisher","first-page":"5410","DOI":"10.1021\/acs.jctc.0c00347","type":"journal-article","volume":"16","author":"Zaverkin","year":"2020","journal-title":"J. Chem. Theory Comput."},{"key":"mlstae09efbib15","doi-asserted-by":"publisher","first-page":"3192","DOI":"10.1039\/C6SC05720A","type":"journal-article","volume":"8","author":"Smith","year":"2017","journal-title":"Chem. Sci."},{"key":"mlstae09efbib16","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.104.136403","type":"journal-article","volume":"104","author":"Bart\u00f3k","year":"2010","journal-title":"Phys. Rev. Lett."},{"key":"mlstae09efbib17","volume":"vol 2","author":"Williams","year":"2006","type":"book"},{"key":"mlstae09efbib18","doi-asserted-by":"publisher","DOI":"10.1063\/1.5051772","type":"journal-article","volume":"149","author":"Uteva","year":"2018","journal-title":"J. Chem. Phys."},{"key":"mlstae09efbib19","doi-asserted-by":"publisher","first-page":"823","DOI":"10.1080\/00268976.2017.1407460","type":"journal-article","volume":"116","author":"Guan","year":"2018","journal-title":"Mol. Phys."},{"key":"mlstae09efbib20","doi-asserted-by":"publisher","first-page":"20","DOI":"10.1038\/s41524-020-0283-z","type":"journal-article","volume":"6","author":"Vandermause","year":"2020","journal-title":"npj Comput. Mater."},{"key":"mlstae09efbib21","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pcbi.1003400","type":"journal-article","volume":"10","author":"Rupp","year":"2014","journal-title":"PLoS Comput. Biol."},{"key":"mlstae09efbib22","first-page":"pp 569","type":"book","article-title":"Bootstrap methods: another look at the jackknife","author":"Efron","year":"1992"},{"key":"mlstae09efbib23","doi-asserted-by":"publisher","DOI":"10.1063\/5.0146905","type":"journal-article","volume":"158","author":"Carrete","year":"2023","journal-title":"J. Chem. Phys."},{"key":"mlstae09efbib24","doi-asserted-by":"publisher","DOI":"10.1088\/2632-2153\/ad8984","type":"journal-article","volume":"5","author":"Christiansen","year":"2024","journal-title":"Mach. Learn.: Sci. Technol."},{"key":"mlstae09efbib25","first-page":"pp 2796","type":"conference-proceedings","article-title":"Accurate uncertainties for deep learning using calibrated regression","author":"Kuleshov","year":"2018"},{"key":"mlstae09efbib26","doi-asserted-by":"publisher","DOI":"10.1063\/5.0109572","type":"journal-article","volume":"157","author":"Pernot","year":"2022","journal-title":"J. Chem. Phys."},{"key":"mlstae09efbib27","doi-asserted-by":"publisher","DOI":"10.1063\/5.0174943","type":"journal-article","volume":"1","author":"Pernot","year":"2023","journal-title":"APL Mach. Learn."},{"key":"mlstae09efbib28","doi-asserted-by":"publisher","first-page":"5540","DOI":"10.3390\/s22155540","type":"journal-article","volume":"22","author":"Levi","year":"2022","journal-title":"Sensors"},{"key":"mlstae09efbib29","doi-asserted-by":"publisher","DOI":"10.1088\/2632-2153\/ab7e1a","type":"journal-article","volume":"1","author":"Tran","year":"2020","journal-title":"Mach. Learn.: Sci. Technol."},{"key":"mlstae09efbib30","doi-asserted-by":"publisher","DOI":"10.1088\/2632-2153\/ad594a","type":"journal-article","volume":"5","author":"Kellner","year":"2024","journal-title":"Mach. Learn.: Sci. Technol."},{"key":"mlstae09efbib31","doi-asserted-by":"publisher","DOI":"10.1063\/5.0136574","type":"journal-article","volume":"158","author":"Zhu","year":"2023","journal-title":"J. Chem. Phys."},{"key":"mlstae09efbib32","doi-asserted-by":"publisher","first-page":"10978","DOI":"10.1039\/C7CP00375G","type":"journal-article","volume":"19","author":"Peterson","year":"2017","journal-title":"Phys. Chem. Chem. Phys."},{"key":"mlstae09efbib33","doi-asserted-by":"publisher","first-page":"25828","DOI":"10.1039\/D3CP02143B","type":"journal-article","volume":"25","author":"Busk","year":"2023","journal-title":"Phys. Chem. Chem. Phys."},{"key":"mlstae09efbib34","first-page":"pp 13","type":"book","article-title":"A sequential algorithm for training text classifiers: Corrigendum and additional data","volume":"vol 29","author":"Lewis","year":"1995"},{"key":"mlstae09efbib35","first-page":"pp 3674","type":"conference-proceedings","article-title":"On the relationship between data efficiency and error for uncertainty sampling","author":"Mussmann","year":"2018"},{"key":"mlstae09efbib36","author":"Bart\u00f3k","year":"2022","type":"preprint"},{"key":"mlstae09efbib37","volume":"vol 31","author":"Gardner","year":"2018","type":"conference-proceedings"},{"key":"mlstae09efbib38","doi-asserted-by":"publisher","DOI":"10.1016\/j.cpc.2019.106949","type":"journal-article","volume":"247","author":"Himanen","year":"2020","journal-title":"Comput. Phys. Commun."},{"key":"mlstae09efbib39","doi-asserted-by":"publisher","first-page":"4012","DOI":"10.1021\/acs.jcim.3c00373","type":"journal-article","volume":"63","author":"Heid","year":"2023","journal-title":"J. Chem. Inf. Model."},{"key":"mlstae09efbib40","first-page":"pp 147","type":"book","volume":"vol 4","author":"Bishop","year":"2006"},{"key":"mlstae09efbib41","author":"Chung","year":"2021","type":"preprint"},{"key":"mlstae09efbib42","doi-asserted-by":"publisher","first-page":"605","DOI":"10.1039\/D2DD00034B","type":"journal-article","volume":"1","author":"Zaverkin","year":"2022","journal-title":"Digit. Discovery"},{"key":"mlstae09efbib43","doi-asserted-by":"publisher","DOI":"10.1016\/j.commatsci.2022.111330","type":"journal-article","volume":"208","author":"Wilson","year":"2022","journal-title":"Comput. Mater. Sci."},{"key":"mlstae09efbib44","doi-asserted-by":"publisher","first-page":"38","DOI":"10.1016\/j.cpc.2019.02.007","type":"journal-article","volume":"240","author":"Chmiela","year":"2019","journal-title":"Comput. Phys. Commun."},{"key":"mlstae09efbib45","doi-asserted-by":"publisher","DOI":"10.1088\/2632-2153\/abba6f","type":"journal-article","volume":"1","author":"Christensen","year":"2020","journal-title":"Mach. Learn.: Sci. Technol."},{"key":"mlstae09efbib46","doi-asserted-by":"publisher","first-page":"95","DOI":"10.1038\/s41597-023-01998-3","type":"journal-article","volume":"10","author":"Pinheiro Jr","year":"2023","journal-title":"Sci. Data"},{"key":"mlstae09efbib47","doi-asserted-by":"publisher","first-page":"1123","DOI":"10.1021\/ed100697w","type":"journal-article","volume":"87","author":"Pence","year":"2010","journal-title":"J. Chem. Educ."},{"key":"mlstae09efbib48","doi-asserted-by":"publisher","first-page":"5678","DOI":"10.1021\/jp070186p","type":"journal-article","volume":"111","author":"Aradi","year":"2007","journal-title":"J. Phys. Chem. A"},{"key":"mlstae09efbib49","doi-asserted-by":"publisher","first-page":"457","DOI":"10.1007\/s10994-021-05946-3","type":"journal-article","volume":"110","author":"H\u00fcllermeier","year":"2021","journal-title":"Mach. Learn."},{"key":"mlstae09efbib50","doi-asserted-by":"publisher","first-page":"171","DOI":"10.1016\/j.commatsci.2017.08.031","type":"journal-article","volume":"140","author":"Podryabinkin","year":"2017","journal-title":"Comput. Mater. Sci."}],"container-title":["Machine Learning: Science and Technology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae09ef","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae09ef\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae09ef","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae09ef\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae09ef\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae09ef\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae09ef\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae09ef\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,27]],"date-time":"2025-10-27T09:06:27Z","timestamp":1761555987000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae09ef"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,27]]},"references-count":50,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2025,10,27]]},"published-print":{"date-parts":[[2025,12,30]]}},"URL":"https:\/\/doi.org\/10.1088\/2632-2153\/ae09ef","relation":{},"ISSN":["2632-2153"],"issn-type":[{"value":"2632-2153","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,10,27]]},"assertion":[{"value":"Evaluation of uncertainty estimations for Gaussian process regression based machine learning interatomic potentials","name":"article_title","label":"Article Title"},{"value":"Machine Learning: Science and Technology","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2025 The Author(s). Published by IOP Publishing Ltd","name":"copyright_information","label":"Copyright Information"},{"value":"2025-01-09","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-09-22","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-10-27","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}