{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,11]],"date-time":"2026-03-11T17:15:25Z","timestamp":1773249325961,"version":"3.50.1"},"reference-count":79,"publisher":"AIP Publishing","issue":"20","license":[{"start":{"date-parts":[[2023,5,22]],"date-time":"2023-05-22T00:00:00Z","timestamp":1684713600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2023,5,22]],"date-time":"2023-05-22T00:00:00Z","timestamp":1684713600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100002428","name":"Austrian Science Fund","doi-asserted-by":"publisher","award":["SFB F81 TACO"],"award-info":[{"award-number":["SFB F81 TACO"]}],"id":[{"id":"10.13039\/501100002428","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100002428","name":"Austrian Science Fund","doi-asserted-by":"publisher","award":["J-4415"],"award-info":[{"award-number":["J-4415"]}],"id":[{"id":"10.13039\/501100002428","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100004837","name":"Ministerio de Ciencia e Innovaci\u00f3n","doi-asserted-by":"publisher","award":["PID2021-126148NA-I00"],"award-info":[{"award-number":["PID2021-126148NA-I00"]}],"id":[{"id":"10.13039\/501100004837","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100015068","name":"Universidade de Santiago de Compostela","doi-asserted-by":"publisher","award":["Convocatoria de Recualificaci\u00f3"],"award-info":[{"award-number":["Convocatoria de Recualificaci\u00f3"]}],"id":[{"id":"10.13039\/501100015068","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100015068","name":"Universidade de Santiago de Compostela","doi-asserted-by":"publisher","award":["n do Sistema Universitario Espa\u00f1"],"award-info":[{"award-number":["n do Sistema Universitario Espa\u00f1"]}],"id":[{"id":"10.13039\/501100015068","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100015068","name":"Universidade de Santiago de Compostela","doi-asserted-by":"publisher","award":["ol- Margarita Salas"],"award-info":[{"award-number":["ol- Margarita Salas"]}],"id":[{"id":"10.13039\/501100015068","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001871","name":"Funda\u00e7\u00e3o para a Ci\u00eancia e a Tecnologia","doi-asserted-by":"publisher","award":["FCT\/MCTES (PIDDAC)"],"award-info":[{"award-number":["FCT\/MCTES (PIDDAC)"]}],"id":[{"id":"10.13039\/501100001871","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["pubs.aip.org"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2023,5,28]]},"abstract":"<jats:p>A reliable uncertainty estimator is a key ingredient in the successful use of machine-learning force fields for predictive calculations. Important considerations are correlation with error, overhead during training and inference, and efficient workflows to systematically improve the force field. However, in the case of neural-network force fields, simple committees are often the only option considered due to their easy implementation. Here, we present a generalization of the deep-ensemble design based on multiheaded neural networks and a heteroscedastic loss. It can efficiently deal with uncertainties in both energy and forces and take sources of aleatoric uncertainty affecting the training data into account. We compare uncertainty metrics based on deep ensembles, committees, and bootstrap-aggregation ensembles using data for an ionic liquid and a perovskite surface. We demonstrate an adversarial approach to active learning to efficiently and progressively refine the force fields. That active learning workflow is realistically possible thanks to exceptionally fast training enabled by residual learning and a nonlinear learned optimizer.<\/jats:p>","DOI":"10.1063\/5.0146905","type":"journal-article","created":{"date-parts":[[2023,5,22]],"date-time":"2023-05-22T10:18:20Z","timestamp":1684750700000},"update-policy":"https:\/\/doi.org\/10.1063\/aip-crossmark-policy-page","source":"Crossref","is-referenced-by-count":51,"title":["Deep ensembles vs committees for uncertainty estimation in neural-network force fields: Comparison and application to active learning"],"prefix":"10.1063","volume":"158","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-0971-1098","authenticated-orcid":false,"given":"Jes\u00fas","family":"Carrete","sequence":"first","affiliation":[{"name":"Institute of Materials Chemistry, TU Wien 1 , A-1060 Vienna, Austria"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9861-8076","authenticated-orcid":false,"given":"Hadri\u00e1n","family":"Montes-Campos","sequence":"additional","affiliation":[{"name":"Grupo de Nanomateriais, Fot\u00f3nica e Materia Branda, Departamento de F\u00edsica de Part\u00edculas, Universidade de Santiago de Compostela 2 , E-15782 Santiago de Compostela, Spain"},{"name":"CIQUP, Institute of Molecular Sciences (IMS)\u2014Departamento de Qu\u00edmica e Bioqu\u00edmica, Faculdade de Ci\u00eancias da Universidade do Porto 3 , Rua Campo Alegre, 4169-007 Porto, Portugal"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3111-9149","authenticated-orcid":false,"given":"Ralf","family":"Wanzenb\u00f6ck","sequence":"additional","affiliation":[{"name":"Institute of Materials Chemistry, TU Wien 1 , A-1060 Vienna, Austria"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8404-6596","authenticated-orcid":false,"given":"Esther","family":"Heid","sequence":"additional","affiliation":[{"name":"Institute of Materials Chemistry, TU Wien 1 , A-1060 Vienna, Austria"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9844-9145","authenticated-orcid":false,"given":"Georg K. H.","family":"Madsen","sequence":"additional","affiliation":[{"name":"Institute of Materials Chemistry, TU Wien 1 , A-1060 Vienna, Austria"}]}],"member":"317","published-online":{"date-parts":[[2023,5,22]]},"reference":[{"key":"2023122821281614700_c1","doi-asserted-by":"publisher","first-page":"4895","DOI":"10.1038\/s41467-020-18556-9","article-title":"Retrospective on a decade of machine learning for chemical discovery","volume":"11","year":"2020","journal-title":"Nat. Commun."},{"key":"2023122821281614700_c2","doi-asserted-by":"publisher","first-page":"1291","DOI":"10.1002\/jcc.24764","article-title":"Deep learning for computational chemistry","volume":"38","year":"2017","journal-title":"J. Comput. Chem."},{"key":"2023122821281614700_c3","doi-asserted-by":"publisher","first-page":"2545","DOI":"10.1021\/acs.jcim.9b00266","article-title":"Deep learning in chemistry","volume":"59","year":"2019","journal-title":"J. Chem. Inf. Model."},{"key":"2023122821281614700_c4","doi-asserted-by":"publisher","first-page":"347","DOI":"10.1038\/s41570-020-0189-9","article-title":"Exploring chemical compound space with quantum-based machine learning","volume":"4","year":"2020","journal-title":"Nat. Rev. Chem."},{"key":"2023122821281614700_c5","doi-asserted-by":"publisher","first-page":"113","DOI":"10.1016\/j.jmst.2020.01.067","article-title":"Machine learning in materials genome initiative: A review","volume":"57","year":"2020","journal-title":"J. Mater. Sci. Technol"},{"key":"2023122821281614700_c6","doi-asserted-by":"publisher","first-page":"513","DOI":"10.1039\/c7sc02664a","article-title":"MoleculeNet: A benchmark for molecular machine learning","volume":"9","year":"2018","journal-title":"Chem. Sci."},{"key":"2023122821281614700_c7","doi-asserted-by":"publisher","first-page":"1241","DOI":"10.1016\/j.drudis.2018.01.039","article-title":"The rise of deep learning in drug discovery","volume":"23","year":"2018","journal-title":"Drug Discov. Today"},{"key":"2023122821281614700_c8","doi-asserted-by":"publisher","first-page":"2594","DOI":"10.1007\/s11095-016-2029-7","article-title":"The next era: Deep learning in pharmaceutical research","volume":"33","year":"2016","journal-title":"Pharm. Res."},{"key":"2023122821281614700_c9","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","year":"2018","journal-title":"Nature"},{"key":"2023122821281614700_c10","doi-asserted-by":"publisher","first-page":"1281","DOI":"10.1021\/acs.accounts.8b00087","article-title":"Machine learning in computer-aided synthesis planning","volume":"51","year":"2018","journal-title":"Acc. Chem. Res."},{"key":"2023122821281614700_c11","doi-asserted-by":"publisher","first-page":"A1133","DOI":"10.1103\/physrev.140.a1133","article-title":"Self-consistent equations including exchange and correlation effects","volume":"140","year":"1965","journal-title":"Phys. Rev."},{"key":"2023122821281614700_c12","doi-asserted-by":"publisher","first-page":"10142","DOI":"10.1021\/acs.chemrev.0c011111","article-title":"Machine learning force fields","volume":"121","year":"2021","journal-title":"Chem. Rev."},{"key":"2023122821281614700_c13","doi-asserted-by":"publisher","first-page":"417","DOI":"10.1007\/s11244-010-9451-6","article-title":"DFT versus the \u2018Real World\u2019 (or, waiting for Godft)","volume":"53","year":"2010","journal-title":"Top. Catal."},{"key":"2023122821281614700_c14","unstructured":"X.\n            Fu\n          , Z.Wu, W.Wang, T.Xie, S.Keten, R.Gomez-Bombarelli, and T.Jaakkola, \u201cForces are not enough: Benchmark and critical evaluation for machine learning force fields with molecular simulations,\u201d arXiv:2210.07237 [physics.comp-ph] (2022)."},{"key":"2023122821281614700_c15","article-title":"International vocabulary of metrology\u2014Basic and general concepts and associated terms (VIM)","author":"BIPM","edition":"3rd ed."},{"key":"2023122821281614700_c16","article-title":"Evaluation of measurement data \u2014 guide to the expression of uncertainty in measurement","author":"BIPM"},{"key":"2023122821281614700_c17","unstructured":"R.\n            Balestriero\n          , J.Pesenti, and Y.LeCun, \u201cLearning in high dimension always amounts to extrapolation,\u201d arXiv:2110.09485 [cs.LG] (2021)."},{"key":"2023122821281614700_c18","doi-asserted-by":"publisher","first-page":"3770","DOI":"10.1021\/acs.jcim.0c00502","article-title":"Uncertainty quantification using neural networks for molecular property prediction","volume":"60","year":"2020","journal-title":"J. Chem. Inf. Model."},{"key":"2023122821281614700_c19","doi-asserted-by":"publisher","first-page":"7913","DOI":"10.1039\/c9sc02298h","article-title":"A quantitative uncertainty metric controls error in neural network-driven chemical discovery","volume":"10","year":"2019","journal-title":"Chem. Sci."},{"key":"2023122821281614700_c20","doi-asserted-by":"publisher","first-page":"025006","DOI":"10.1088\/2632-2153\/ab7e1a","article-title":"Methods for comparing uncertainty quantifications for material property predictions","volume":"1","year":"2020","journal-title":"Mach. Learn.: Sci. Technol."},{"key":"2023122821281614700_c21","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1214\/aos\/1176344552","article-title":"Bootstrap methods: Another look at the Jackknife","volume":"7","year":"1979","journal-title":"Ann. Stat."},{"key":"2023122821281614700_c22","first-page":"1050","article-title":"Dropout as a Bayesian approximation: Representing model uncertainty in deep learning","year":"2016"},{"key":"2023122821281614700_c23","doi-asserted-by":"publisher","first-page":"2031","DOI":"10.1214\/aos\/1176325770","article-title":"Large sample confidence regions based on subsamples under minimal assumptions","volume":"22","year":"1994","journal-title":"Ann. Stat."},{"key":"2023122821281614700_c24","doi-asserted-by":"publisher","first-page":"55","DOI":"10.1109\/icnn.1994.374138","article-title":"Estimating the mean and variance of the target probability distribution","year":"1994"},{"key":"2023122821281614700_c25","doi-asserted-by":"publisher","first-page":"1356","DOI":"10.1021\/acscentsci.1c00546","article-title":"Evidential deep learning for guided molecular property prediction and discovery","volume":"7","year":"2021","journal-title":"ACS Cent. Sci."},{"key":"2023122821281614700_c26","doi-asserted-by":"crossref","unstructured":"E.\n            Heid\n          , C. J.McGill, F. H.Vermeire, and W. H.Green, \u201cCharacterizing uncertainty in machine learning for chemistry,\u201dchemRxiv:10.26434 (2023).10.26434\/chemrxiv-2023-00vcg-v2","DOI":"10.26434\/chemrxiv-2023-00vcg-v2"},{"key":"2023122821281614700_c27","article-title":"Simple and scalable predictive uncertainty estimation using deep ensembles","year":"2017"},{"key":"2023122821281614700_c28","doi-asserted-by":"publisher","first-page":"146401","DOI":"10.1103\/physrevlett.98.146401","article-title":"Generalized neural-network representation of high-dimensional potential-energy surfaces","volume":"98","year":"2007","journal-title":"Phys. Rev. Lett."},{"key":"2023122821281614700_c29","doi-asserted-by":"publisher","first-page":"241708","DOI":"10.1063\/1.5017898","article-title":"A reactive, scalable, and transferable model for molecular energies from a neural network approach based on local information","volume":"148","year":"2018","journal-title":"J. Chem. Phys."},{"key":"2023122821281614700_c30","doi-asserted-by":"publisher","first-page":"2453","DOI":"10.1038\/s41467-022-29939-5","article-title":"E(3)-equivariant graph neural networks for data-efficient and accurate interatomic potentials","volume":"13","year":"2022","journal-title":"Nat. Commun."},{"key":"2023122821281614700_c31","unstructured":"Y.-L.\n            Liao\n           and T.Smidt, \u201cEquiformer: Equivariant graph attention transformer for 3d atomistic graphs,\u201d arXiv:2206.11990 [cs.LG] (2022)."},{"key":"2023122821281614700_c32","doi-asserted-by":"publisher","first-page":"e1603015","DOI":"10.1126\/sciadv.1603015","article-title":"Machine learning of accurate energy-conserving molecular force fields","volume":"3","year":"2017","journal-title":"Sci. Adv."},{"key":"2023122821281614700_c33","doi-asserted-by":"publisher","first-page":"184115","DOI":"10.1103\/physrevb.87.184115","article-title":"On representing chemical environments","volume":"87","year":"2013","journal-title":"Phys. Rev. B"},{"key":"2023122821281614700_c34","first-page":"1263","article-title":"Neural message passing for quantum chemistry","year":"2017"},{"key":"2023122821281614700_c35","article-title":"SchNet: A continuous-filter convolutional neural network for modeling quantum interactions","year":"2017"},{"key":"2023122821281614700_c36","unstructured":"J.\n            Gasteiger\n          , J.Gro\u00df, and S.G\u00fcnnemann, \u201cDirectional message passing for molecular graphs,\u201d arXiv:2003.03123 [cs.LG] (2020)."},{"key":"2023122821281614700_c37","doi-asserted-by":"publisher","first-page":"045439","DOI":"10.1103\/physrevb.85.045439","article-title":"High-dimensional neural network potentials for metal surfaces: A prototype study for copper","volume":"85","year":"2012","journal-title":"Phys. Rev. B"},{"key":"2023122821281614700_c38","doi-asserted-by":"publisher","first-page":"241733","DOI":"10.1063\/1.5023802","article-title":"Less is more: Sampling chemical space with active learning","volume":"148","year":"2018","journal-title":"J. Chem. Phys."},{"key":"2023122821281614700_c39","doi-asserted-by":"publisher","first-page":"124113","DOI":"10.1063\/1.5019675","article-title":"Reinforced dynamics for enhanced sampling in large atomic and molecular systems","volume":"148","year":"2018","journal-title":"J. Chem. Phys."},{"key":"2023122821281614700_c40","doi-asserted-by":"publisher","first-page":"906","DOI":"10.1021\/acs.jctc.8b00959","article-title":"Fast and accurate uncertainty estimation in chemical machine learning","volume":"15","year":"2019","journal-title":"J. Chem. Theory Comput."},{"key":"2023122821281614700_c41","doi-asserted-by":"publisher","first-page":"104105","DOI":"10.1063\/5.0016004","article-title":"Committee neural network potentials control generalization errors and enable active learning","volume":"153","year":"2020","journal-title":"J. Chem. Phys."},{"key":"2023122821281614700_c42","doi-asserted-by":"publisher","first-page":"88","DOI":"10.1021\/acs.jcim.1c01380","article-title":"A differentiable neural-network force field for ionic liquids","volume":"62","year":"2022","journal-title":"J. Chem. Inf. Model."},{"key":"2023122821281614700_c43","unstructured":"W. M.\n            Czarnecki\n          , S.Osindero, M.Jaderberg, G.Swirszcz, and R.Pascanu, \u201cSobolev training for neural networks,\u201d arXiv:1706.04859 [cs.LG] (2017)."},{"key":"2023122821281614700_c44","doi-asserted-by":"publisher","first-page":"916","DOI":"10.1109\/tnnls.2020.2979706","article-title":"Enhancing function approximation abilities of neural networks by training derivatives","volume":"32","year":"2021","journal-title":"IEEE Trans. Neural Networks Learn. Syst."},{"key":"2023122821281614700_c45","doi-asserted-by":"publisher","first-page":"770","DOI":"10.1109\/cvpr.2016.90","article-title":"Deep residual learning for image recognition","year":"2016"},{"key":"2023122821281614700_c46","unstructured":"L.\n            Metz\n          , J.Harrison, C. D.Freeman, A.Merchant, L.Beyer, J.Bradbury, N.Agrawal, B.Poole, I.Mordatch, A.Roberts, and J.Sohl-Dickstein, \u201cVeLO: Training versatile learned optimizers by scaling up,\u201d arXiv:2211.09760 [cs.LG] (2022)."},{"key":"2023122821281614700_c47","doi-asserted-by":"publisher","first-page":"5104","DOI":"10.1038\/s41467-021-25342-8","article-title":"Differentiable sampling of molecular geometries with uncertainty-based adversarial attacks","volume":"12","year":"2021","journal-title":"Nat. Commun."},{"key":"2023122821281614700_c48","doi-asserted-by":"publisher","first-page":"015021","DOI":"10.1063\/1.5111045","article-title":"Continuous and optimally complete description of chemical environments using spherical Bessel descriptors","volume":"10","year":"2020","journal-title":"AIP Adv."},{"key":"2023122821281614700_c49","article-title":"Deep sets","volume-title":"Advances in Neural Information Processing Systems","author":"Guyon","year":"2017"},{"key":"2023122821281614700_c50","first-page":"6487","article-title":"On the limitations of representing functions on sets","volume-title":"Proceedings of Machine Learning Research","author":"Chaudhuri","year":"2019"},{"key":"2023122821281614700_c51","article-title":"On the representation power of set pooling networks","volume-title":"Adv. Neural. Inf. Process. Syst","author":"Beygelzimer","year":"2021"},{"key":"2023122821281614700_c52","unstructured":"S.\n            Bartunov\n          , F. B.Fuchs, and T.Lillicrap, \u201cEquilibrium aggregation: Encoding sets via optimization,\u201d arXiv:2202.12795 [cs.LG] (2022)."},{"key":"2023122821281614700_c53","doi-asserted-by":"publisher","first-page":"79","DOI":"10.1007\/s10909-016-1617-6","article-title":"Gauge invariance of thermal transport coefficients","volume":"185","year":"2016","journal-title":"J. Low Temp. Phys."},{"key":"2023122821281614700_c54","unstructured":"P.\n            Ramachandran\n          , B.Zoph, and Q. V.Le, \u201cSearching for activation functions,\u201d arXiv:1710.05941 [cs.NE] (2017)."},{"key":"2023122821281614700_c55","unstructured":"J. L.\n            Ba\n          , J. R.Kiros, and G. E.Hinton, \u201cLayer normalization,\u201d arXiv:1607.06450 [stat.ML] (2016)."},{"key":"2023122821281614700_c56","doi-asserted-by":"publisher","first-page":"193","DOI":"10.3390\/e22020193","article-title":"Deep residual learning for nonlinear regression","volume":"22","year":"2020","journal-title":"Entropy"},{"key":"2023122821281614700_c57","unstructured":"S.\n            Ioffe\n           and C.Szegedy, \u201cBatch normalization: Accelerating deep network training by reducing internal covariate shift,\u201d arXiv:1502.03167[cs.LG] (2015)."},{"key":"2023122821281614700_c58","doi-asserted-by":"publisher","first-page":"e27095","DOI":"10.1002\/qua.27095","article-title":"Evaluating the efficiency of power-series expansions as model potentials for finite-temperature atomistic calculations","volume":"123","year":"2023","journal-title":"Int. J. Quantum Chem."},{"key":"2023122821281614700_c59","volume-title":"An Introduction to Manifolds","year":"2010"},{"key":"2023122821281614700_c60","article-title":"JAX: Composable transformations of Python+NumPy programs","year":"2018"},{"key":"2023122821281614700_c61","article-title":"Flax: A neural network library and ecosystem for JAX","year":"2020"},{"key":"2023122821281614700_c62","doi-asserted-by":"publisher","article-title":"Deep ensembles vs. committees for uncertainty estimation in neural-network force fields: Comparison and application to active learning","DOI":"10.5281\/zenodo.7643625"},{"key":"2023122821281614700_c63","doi-asserted-by":"publisher","first-page":"9","DOI":"10.1007\/978-3-642-35289-8_3","article-title":"Efficient backprop","volume-title":"Neural Networks: Tricks of the Trade","author":"Montavon","year":"2012","edition":"2nd ed."},{"key":"2023122821281614700_c64","volume-title":"Generalized Hyperbolic Secant Distributions","year":"2013","edition":"2014th ed."},{"key":"2023122821281614700_c65","article-title":"Adam: A method for stochastic optimization"},{"key":"2023122821281614700_c66","article-title":"The DeepMind JAX ecosystem","year":"2020"},{"key":"2023122821281614700_c67","doi-asserted-by":"publisher","first-page":"164111","DOI":"10.1063\/5.0136574","article-title":"Fast uncertainty estimates in deep learning interatomic potentials","volume":"158","year":"2023","journal-title":"J. Chem. Phys."},{"key":"2023122821281614700_c68","doi-asserted-by":"crossref","unstructured":"R.\n            Egele\n          , R.Maulik, K.Raghavan, P.Balaprakash, and B.Lusch, \u201cAutoDEUQ: Automated deep ensemble with uncertainty quantification,\u201d arXiv:2110.13511 [stat.ML] (2021).","DOI":"10.1109\/ICPR56361.2022.9956231"},{"key":"2023122821281614700_c69","doi-asserted-by":"publisher","first-page":"035109","DOI":"10.1103\/physrevb.71.035109","article-title":"Real-space grid implementation of the projector augmented wave method","volume":"71","year":"2005","journal-title":"Phys. Rev. B"},{"key":"2023122821281614700_c70","doi-asserted-by":"publisher","first-page":"253202","DOI":"10.1088\/0953-8984\/22\/25\/253202","article-title":"Electronic structure calculations with GPAW: A real-space implementation of the projector augmented-wave method","volume":"22","year":"2010","journal-title":"J. Phys.: Condens. Matter"},{"key":"2023122821281614700_c71","article-title":"JAX, M.D.: A framework for differentiable physics","volume-title":"Advances in Neural Information Processing Systems","year":"2020"},{"key":"2023122821281614700_c72","doi-asserted-by":"publisher","first-page":"5629","DOI":"10.1088\/0305-4470\/39\/19\/s18","article-title":"A Liouville-operator derived measure-preserving integrator for molecular dynamics simulations in the isothermal\u2013isobaric ensemble","volume":"39","year":"2006","journal-title":"J. Phys. A: Math. Theor."},{"key":"2023122821281614700_c73","doi-asserted-by":"publisher","first-page":"115122","DOI":"10.1103\/physrevb.73.115122","article-title":"Filtering a distribution simultaneously in real and Fourier space","volume":"73","year":"2006","journal-title":"Phys. Rev. B"},{"key":"2023122821281614700_c74","doi-asserted-by":"publisher","first-page":"703","DOI":"10.1039\/d2dd00072e","article-title":"Neural-network-backed evolutionary search for SrTiO3(110) surface reconstructions","volume":"1","year":"2022","journal-title":"Digit. Discov."},{"key":"2023122821281614700_c75","doi-asserted-by":"publisher","first-page":"022001","DOI":"10.1103\/physrevmaterials.3.022001","article-title":"Low-temperature specific heat of doped SrTiO3: Doping dependence of the effective mass and Kadowaki-Woods scaling violation","volume":"3","year":"2019","journal-title":"Phys. Rev. Mater."},{"key":"2023122821281614700_c76","doi-asserted-by":"publisher","first-page":"3865","DOI":"10.1103\/physrevlett.77.3865","article-title":"Generalized gradient approximation made simple","volume":"77","year":"1996","journal-title":"Phys. Rev. Lett."},{"key":"2023122821281614700_c77","doi-asserted-by":"publisher","first-page":"155","DOI":"10.1093\/comjnl\/7.2.155","article-title":"An efficient method for finding the minimum of a function of several variables without calculating derivatives","volume":"7","year":"1964","journal-title":"Comput. J."},{"key":"2023122821281614700_c78","doi-asserted-by":"publisher","first-page":"023804","DOI":"10.1103\/physrevmaterials.3.023804","article-title":"Active learning of uniformly accurate interatomic potentials for materials simulation","volume":"3","year":"2019","journal-title":"Phys. Rev. Mater."},{"key":"2023122821281614700_c79","doi-asserted-by":"publisher","first-page":"3846","DOI":"10.1021\/acs.jcim.1c00670","article-title":"Pairwise difference regression: A machine learning meta-algorithm for improved prediction and uncertainty quantification in chemical search","volume":"61","year":"2021","journal-title":"J. Chem. Inf. Model."}],"container-title":["The Journal of Chemical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.aip.org\/aip\/jcp\/article-pdf\/doi\/10.1063\/5.0146905\/18281408\/204801_1_5.0146905.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/pubs.aip.org\/aip\/jcp\/article-pdf\/doi\/10.1063\/5.0146905\/18281408\/204801_1_5.0146905.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.aip.org\/aip\/jcp\/article-pdf\/doi\/10.1063\/5.0146905\/18281408\/204801_1_5.0146905.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,5,27]],"date-time":"2025-05-27T20:20:43Z","timestamp":1748377243000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.aip.org\/jcp\/article\/158\/20\/204801\/2891492\/Deep-ensembles-vs-committees-for-uncertainty"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,5,22]]},"references-count":79,"journal-issue":{"issue":"20","published-print":{"date-parts":[[2023,5,28]]}},"URL":"https:\/\/doi.org\/10.1063\/5.0146905","relation":{},"ISSN":["0021-9606","1089-7690"],"issn-type":[{"value":"0021-9606","type":"print"},{"value":"1089-7690","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2023,5,28]]},"published":{"date-parts":[[2023,5,22]]},"article-number":"204801"}}