{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,3]],"date-time":"2025-11-03T09:22:42Z","timestamp":1762161762659,"version":"3.37.3"},"reference-count":64,"publisher":"IOP Publishing","issue":"3","license":[{"start":{"date-parts":[[2022,9,22]],"date-time":"2022-09-22T00:00:00Z","timestamp":1663804800000},"content-version":"vor","delay-in-days":21,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2022,9,22]],"date-time":"2022-09-22T00:00:00Z","timestamp":1663804800000},"content-version":"tdm","delay-in-days":21,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"DOI":"10.13039\/501100001711","name":"Schweizerischer Nationalfonds zur F\u00f6rderung der Wissenschaftlichen Forschung","doi-asserted-by":"crossref","award":["185030"],"award-info":[{"award-number":["185030"]}],"id":[{"id":"10.13039\/501100001711","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/100010663","name":"H2020 European Research Council","doi-asserted-by":"crossref","award":["817977"],"award-info":[{"award-number":["817977"]}],"id":[{"id":"10.13039\/100010663","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":[[2022,9,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Supervised and unsupervised kernel-based algorithms widely used in the physical sciences depend upon the notion of <jats:italic>similarity<\/jats:italic>. Their reliance on pre-defined distance metrics\u2014e.g. the Euclidean or Manhattan distance\u2014are problematic especially when used in combination with high-dimensional feature vectors for which the similarity measure does not well-reflect the differences in the target property. <jats:italic>Metric learning<\/jats:italic> is an elegant approach to surmount this shortcoming and find a property-informed transformation of the feature space. We propose a new algorithm for metric learning specifically adapted for kernel ridge regression (KRR): <jats:italic>metric learning for kernel ridge regression<\/jats:italic> (MLKRR). It is based on the Metric Learning for Kernel Regression framework using the Nadaraya-Watson estimator, which we show to be inferior to the KRR estimator for typical physics-based machine learning tasks. The MLKRR algorithm allows for superior predictive performance on the benchmark regression task of atomisation energies of QM9 molecules, as well as generating more meaningful low-dimensional projections of the modified feature space.<\/jats:p>","DOI":"10.1088\/2632-2153\/ac8e4f","type":"journal-article","created":{"date-parts":[[2022,8,31]],"date-time":"2022-08-31T22:42:43Z","timestamp":1661985763000},"page":"035015","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":10,"title":["Metric learning for kernel ridge regression: assessment of molecular similarity"],"prefix":"10.1088","volume":"3","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-7946-817X","authenticated-orcid":true,"given":"Raimon","family":"Fabregat","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7992-5529","authenticated-orcid":false,"given":"Puck","family":"van Gerwen","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Matthieu","family":"Haeberle","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Friedrich","family":"Eisenbrand","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7993-2879","authenticated-orcid":true,"given":"Cl\u00e9mence","family":"Corminboeuf","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"266","published-online":{"date-parts":[[2022,9,22]]},"reference":[{"key":"mlstac8e4fbib1","doi-asserted-by":"publisher","DOI":"10.1088\/2632-2153\/ab6d5d","article-title":"Introducing machine learning: science and technology","volume":"1","author":"von Lilienfeld","year":"2020","journal-title":"Mach. Learn.: Sci. Technol."},{"key":"mlstac8e4fbib2","doi-asserted-by":"publisher","first-page":"e8","DOI":"10.1002\/ail2.8","article-title":"Welcome to the first issue of applied AI letters","volume":"1","author":"Pyzer-Knapp","year":"2020","journal-title":"Appl. AI lett."},{"key":"mlstac8e4fbib3","doi-asserted-by":"publisher","first-page":"10001","DOI":"10.1021\/acs.chemrev.0c01303","article-title":"Ab initio machine learning in chemical compound space","volume":"121","author":"Huang","year":"2021","journal-title":"Chem. Rev."},{"key":"mlstac8e4fbib4","doi-asserted-by":"publisher","first-page":"9759","DOI":"10.1021\/acs.chemrev.1c00021","article-title":"Physics-inspired structural representations for molecules and materials","volume":"121","author":"Musil","year":"2021","journal-title":"Chem. Rev."},{"key":"mlstac8e4fbib5","doi-asserted-by":"publisher","first-page":"547","DOI":"10.1038\/s41586-018-0337-2","article-title":"Machine learning for molecular and materials science","volume":"559","author":"Butler","year":"2018","journal-title":"Nature"},{"key":"mlstac8e4fbib6","doi-asserted-by":"publisher","first-page":"10142","DOI":"10.1021\/acs.chemrev.0c01111","article-title":"Machine learning force fields","volume":"121","author":"Unke","year":"2021","journal-title":"Chem. Rev."},{"key":"mlstac8e4fbib7","doi-asserted-by":"publisher","first-page":"144","DOI":"10.1021\/acscentsci.7b00550","article-title":"The matter simulation (R)evolution","volume":"4","author":"Aspuru-Guzik","year":"2018","journal-title":"ACS Cent. Sci."},{"key":"mlstac8e4fbib8","doi-asserted-by":"publisher","first-page":"230","DOI":"10.1038\/s41929-018-0056-y","article-title":"Machine learning in catalysis","volume":"1","author":"Kitchin","year":"2018","journal-title":"Nat. Catal."},{"key":"mlstac8e4fbib9","doi-asserted-by":"publisher","DOI":"10.1103\/RevModPhys.91.045002","article-title":"Machine learning and the physical sciences","volume":"91","author":"Carleo","year":"2019","journal-title":"Rev. Mod. Phys."},{"key":"mlstac8e4fbib10","doi-asserted-by":"publisher","DOI":"10.1063\/1.5091842","article-title":"Unsupervised machine learning in atomistic simulations, between predictions and understanding","volume":"150","author":"Ceriotti","year":"2019","journal-title":"J. Chem. Phys."},{"key":"mlstac8e4fbib11","doi-asserted-by":"publisher","first-page":"9722","DOI":"10.1021\/acs.chemrev.0c01195","article-title":"Unsupervised learning methods for molecular simulation data","volume":"121","author":"Glielmo","year":"2021","journal-title":"Chem. Rev."},{"key":"mlstac8e4fbib12","doi-asserted-by":"publisher","first-page":"1981","DOI":"10.1021\/acs.accounts.0c00403","article-title":"Mapping materials and molecules","volume":"53","author":"Cheng","year":"2020","journal-title":"Acc. Chem. Res."},{"key":"mlstac8e4fbib13","doi-asserted-by":"publisher","first-page":"10073","DOI":"10.1021\/acs.chemrev.1c00022","article-title":"Gaussian process regression for materials and molecules","volume":"121","author":"Deringer","year":"2021","journal-title":"Chem. Rev."},{"key":"mlstac8e4fbib14","doi-asserted-by":"publisher","DOI":"10.1063\/1.5003074","article-title":"Neural networks vs Gaussian process regression for representing potential energy surfaces: a comparative study of fit quality and vibrational spectrum accuracy","volume":"148","author":"Kamath","year":"2018","journal-title":"J. Chem. Phys."},{"key":"mlstac8e4fbib15","doi-asserted-by":"publisher","first-page":"5255","DOI":"10.1021\/acs.jctc.7b00577","article-title":"Prediction errors of molecular machine learning models lower than hybrid DFT error","volume":"13","author":"Faber","year":"2017","journal-title":"J. Chem. Theory Comput."},{"article-title":"Directional message passing for molecular graphs","year":"2020","author":"Klicpera","key":"mlstac8e4fbib16"},{"key":"mlstac8e4fbib17","article-title":"Cormorant: covariant molecular neural networks","volume":"vol 32","author":"Anderson","year":"2019"},{"key":"mlstac8e4fbib18","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41467-022-29939-5","article-title":"E(3)-equivariant graph neural networks for data-efficient and accurate interatomic potentials","volume":"13","author":"Batzner","year":"2022","journal-title":"Nat. Commun."},{"key":"mlstac8e4fbib19","first-page":"9323","article-title":"E(n) equivariant graph neural networks","volume":"vol 139","author":"Satorras","year":"2021"},{"key":"mlstac8e4fbib20","first-page":"2579","article-title":"Visualizing data using t-SNE","volume":"9","author":"Van der Maaten","year":"2008","journal-title":"J. Mach. Learn. Res."},{"key":"mlstac8e4fbib21","doi-asserted-by":"publisher","first-page":"2319","DOI":"10.1126\/science.290.5500.2319","article-title":"A global geometric framework for nonlinear dimensionality reduction","volume":"290","author":"Tenenbaum","year":"2000","journal-title":"Science"},{"key":"mlstac8e4fbib22","doi-asserted-by":"publisher","first-page":"945","DOI":"10.1038\/s41557-020-0527-z","article-title":"Quantum machine learning using atom-in-molecule-based fragments selected on the fly","volume":"12","author":"Huang","year":"2020","journal-title":"Nat. Chem."},{"key":"mlstac8e4fbib23","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevB.87.184115","article-title":"On representing chemical environments","volume":"87","author":"Bart\u00f3k","year":"2013","journal-title":"Phys. Rev. B"},{"key":"mlstac8e4fbib24","doi-asserted-by":"publisher","DOI":"10.1063\/1.5020710","article-title":"Alchemical and structural distribution based representation for universal quantum machine learning","volume":"148","author":"Faber","year":"2018","journal-title":"J. Chem. Phys."},{"key":"mlstac8e4fbib25","doi-asserted-by":"publisher","DOI":"10.1063\/1.5126701","article-title":"FCHL revisited: faster and more accurate quantum machine learning","volume":"152","author":"Christensen","year":"2020","journal-title":"J. Chem. Phys."},{"key":"mlstac8e4fbib26","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.108.058301","article-title":"Fast and accurate modeling of molecular atomization energies with machine learning","volume":"108","author":"Rupp","year":"2012","journal-title":"Phys. Rev. Lett."},{"key":"mlstac8e4fbib27","doi-asserted-by":"publisher","DOI":"10.1063\/1.4964627","article-title":"Communication: understanding molecular representations in machine learning: the role of uniqueness and target similarity","volume":"145","author":"Huang","year":"2016","journal-title":"J. Chem. Phys."},{"key":"mlstac8e4fbib28","doi-asserted-by":"publisher","first-page":"4164","DOI":"10.1002\/anie.201709686","article-title":"Quantum machine learning in chemical compound space","volume":"57","author":"Von Lilienfeld","year":"2018","journal-title":"Angew. Chem., Int. Ed."},{"key":"mlstac8e4fbib29","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.114.096405","article-title":"Molecular dynamics with on-the-fly machine learning of quantum-mechanical forces","volume":"114","author":"Li","year":"2015","journal-title":"Phys. Rev. Lett."},{"key":"mlstac8e4fbib30","doi-asserted-by":"publisher","DOI":"10.1126\/sciadv.1603015","article-title":"Machine learning of accurate energy-conserving molecular force fields","volume":"3","author":"Chmiela","year":"2017","journal-title":"Sci. Adv."},{"key":"mlstac8e4fbib31","doi-asserted-by":"publisher","first-page":"3887","DOI":"10.1038\/s41467-018-06169-2","article-title":"Towards exact molecular dynamics simulations with machine-learned force fields","volume":"9","author":"Chmiela","year":"2018","journal-title":"Nat. Commun."},{"key":"mlstac8e4fbib32","doi-asserted-by":"publisher","first-page":"3225","DOI":"10.1021\/acs.jctc.5b00301","article-title":"Transferable atomic multipole machine learning models for small organic molecules","volume":"11","author":"Bereau","year":"2015","journal-title":"J. Chem. Theory Comput."},{"key":"mlstac8e4fbib33","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.120.036002","article-title":"Symmetry-adapted machine learning for tensorial properties of atomistic systems","volume":"120","author":"Grisafi","year":"2018","journal-title":"Phys. Rev. Lett."},{"key":"mlstac8e4fbib34","doi-asserted-by":"publisher","first-page":"3401","DOI":"10.1073\/pnas.1816132116","article-title":"Accurate molecular polarizabilities with coupled cluster theory and machine learning","volume":"116","author":"Wilkins","year":"2019","journal-title":"Proc. Natl Acad. Sci. USA"},{"key":"mlstac8e4fbib35","doi-asserted-by":"publisher","first-page":"57","DOI":"10.1021\/acscentsci.8b00551","article-title":"Transferable machine-learning model of the electron density","volume":"5","author":"Grisafi","year":"2018","journal-title":"ACS Cent. Sci."},{"key":"mlstac8e4fbib36","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.104.136403","article-title":"Gaussian approximation potentials: the accuracy of quantum mechanics, without the electrons","volume":"104","author":"Bart\u00f3k","year":"2010","journal-title":"Phys. Rev. Lett."},{"key":"mlstac8e4fbib37","doi-asserted-by":"publisher","first-page":"9424","DOI":"10.1039\/C9SC02696G","article-title":"Electron density learning of non-covalent systems","volume":"10","author":"Fabrizio","year":"2019","journal-title":"Chem. Sci."},{"key":"mlstac8e4fbib38","doi-asserted-by":"publisher","first-page":"8100","DOI":"10.1039\/C9SC01742A","article-title":"Machine learning enables long time scale molecular photodynamics simulations","volume":"10","author":"Westermayr","year":"2019","journal-title":"Chem. Sci."},{"key":"mlstac8e4fbib39","doi-asserted-by":"publisher","first-page":"6879","DOI":"10.1039\/D1SC00482D","article-title":"Reaction-based machine learning representations for predicting the enantioselectivity of organocatalysts","volume":"12","author":"Gallarati","year":"2021","journal-title":"Chem. Sci."},{"key":"mlstac8e4fbib40","doi-asserted-by":"publisher","first-page":"697","DOI":"10.1073\/pnas.0803205106","article-title":"CUR matrix decompositions for improved data analysis","volume":"106","author":"Mahoney","year":"2009","journal-title":"Proc. Natl Acad. Sci. USA"},{"key":"mlstac8e4fbib41","doi-asserted-by":"publisher","DOI":"10.1007\/978-1-4614-6849-3","author":"Kuhn","year":"2013","journal-title":"Applied Predictive Modelling"},{"key":"mlstac8e4fbib42","doi-asserted-by":"publisher","first-page":"287","DOI":"10.1561\/2200000019","article-title":"Metric learning: a survey","volume":"5","author":"Kulis","year":"2013","journal-title":"Found. Trends Mach. Learn."},{"year":"2006","author":"Yang","key":"mlstac8e4fbib43"},{"key":"mlstac8e4fbib44","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s13321-022-00595-7","article-title":"Application of deep metric learning to molecular graph similarity","volume":"14","author":"Coupry","year":"2022","journal-title":"J. Cheminformatics"},{"key":"mlstac8e4fbib45","first-page":"612","article-title":"Metric learning for kernel regression","volume":"vol 2","author":"Weinberger","year":"2007","edition":"ed"},{"key":"mlstac8e4fbib46","article-title":"Siamese neural networks for one-shot image recognition","volume":"vol 37","author":"Koch","year":"2015"},{"key":"mlstac8e4fbib47","first-page":"84","article-title":"Deep metric learning using triplet network","volume":"vol 9370","author":"Hoffer","year":"2015","edition":"ed"},{"year":"2006","author":"Rasmussen","key":"mlstac8e4fbib48"},{"key":"mlstac8e4fbib49","doi-asserted-by":"publisher","first-page":"539","DOI":"10.1109\/CVPR.2005.202","article-title":"Learning a similarity metric discriminatively, with application to face verification","volume":"vol 1","author":"Chopra","year":"2005"},{"key":"mlstac8e4fbib50","first-page":"4004","article-title":"Deep metric learning via lifted structured feature embedding","author":"Oh Song","year":"2016"},{"key":"mlstac8e4fbib51","first-page":"18661","article-title":"Supervised contrastive learning","volume":"vol 33","author":"Khosla","year":"2020","edition":"ed"},{"key":"mlstac8e4fbib52","first-page":"20479","article-title":"3D infomax improves GNNs for molecular property prediction","volume":"vol 162","author":"St\u00e4rk","year":"2022"},{"key":"mlstac8e4fbib53","doi-asserted-by":"publisher","first-page":"473","DOI":"10.1145\/1273496.1273556","article-title":"An empirical evaluation of deep architectures on problems with many factors of variation","author":"Larochelle","year":"2007"},{"key":"mlstac8e4fbib54","doi-asserted-by":"publisher","first-page":"128","DOI":"10.3390\/molecules26010128","article-title":"Improved deep learning based method for molecular similarity searching using stack of deep belief networks","volume":"26","author":"Nasser","year":"2020","journal-title":"Molecules"},{"key":"mlstac8e4fbib55","first-page":"3242","article-title":"Beyond similar and dissimilar relations: a kernel regression formulation for metric learning","author":"Zhu","year":"2018"},{"key":"mlstac8e4fbib56","doi-asserted-by":"publisher","DOI":"10.1038\/sdata.2014.22","article-title":"Quantum chemistry structures and properties of 134 kilo molecules","volume":"1","author":"Ramakrishnan","year":"2014","journal-title":"Sci. Data"},{"key":"mlstac8e4fbib57","first-page":"49","article-title":"On the generalized distance in statistics","volume":"2","author":"Mahalanobis","year":"1936","journal-title":"Proc. Indian National Sci. Acad."},{"year":"2019","author":"Welling","key":"mlstac8e4fbib58"},{"key":"mlstac8e4fbib59","first-page":"1","article-title":"Metric-learn: metric learning algorithms in Python","volume":"21","author":"De Vazelhes","year":"2020","journal-title":"J. Mach. Learn. Res."},{"article-title":"QML: a Python toolkit for quantum machine learning","year":"2017","author":"Christensen","key":"mlstac8e4fbib60"},{"key":"mlstac8e4fbib61","doi-asserted-by":"publisher","first-page":"261","DOI":"10.1038\/s41592-019-0686-2","article-title":"SciPy 1.0: fundamental algorithms for scientific computing in Python","volume":"17","author":"Virtanen","year":"2020","journal-title":"Nat. Methods"},{"year":"2004","author":"Boyd","key":"mlstac8e4fbib62"},{"key":"mlstac8e4fbib63","doi-asserted-by":"publisher","first-page":"29661","DOI":"10.1039\/C8CP05921G","article-title":"Feature optimization for atomistic machine learning yields a data-driven construction of the periodic table of the elements","volume":"20","author":"Willatt","year":"2018","journal-title":"Phys. Chem. Chem. Phys."},{"key":"mlstac8e4fbib64","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-540-73263-1","author":"Brazdil","year":"2009","edition":"ed"}],"container-title":["Machine Learning: Science and Technology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac8e4f","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac8e4f\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac8e4f","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac8e4f\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac8e4f\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac8e4f\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac8e4f\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac8e4f\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,9,22]],"date-time":"2022-09-22T07:43:02Z","timestamp":1663832582000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac8e4f"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,9,1]]},"references-count":64,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2022,9,22]]},"published-print":{"date-parts":[[2022,9,1]]}},"URL":"https:\/\/doi.org\/10.1088\/2632-2153\/ac8e4f","relation":{},"ISSN":["2632-2153"],"issn-type":[{"type":"electronic","value":"2632-2153"}],"subject":[],"published":{"date-parts":[[2022,9,1]]},"assertion":[{"value":"Metric learning for kernel ridge regression: assessment of molecular similarity","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 2022 The Author(s). Published by IOP Publishing Ltd","name":"copyright_information","label":"Copyright Information"},{"value":"2022-06-20","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2022-08-31","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2022-09-22","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}