{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,18]],"date-time":"2025-11-18T09:45:20Z","timestamp":1763459120627,"version":"3.45.0"},"reference-count":88,"publisher":"IOP Publishing","issue":"4","license":[{"start":{"date-parts":[[2025,11,18]],"date-time":"2025-11-18T00:00:00Z","timestamp":1763424000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2025,11,18]],"date-time":"2025-11-18T00:00:00Z","timestamp":1763424000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"name":"Danmarks Grundforskningsfond","award":["150"],"award-info":[{"award-number":["150"]}]},{"DOI":"10.13039\/100008398","name":"Villum Fonden","doi-asserted-by":"crossref","award":["16562"],"award-info":[{"award-number":["16562"]}],"id":[{"id":"10.13039\/100008398","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>Machine learning interatomic potentials have become an indispensable tool for materials science, enabling the study of larger systems and longer timescales. State-of-the-art models are generally graph neural networks that employ message passing to iteratively update atomic embeddings that are ultimately used for predicting properties. In this work we extend the message passing formalism with the inclusion of a continuous variable that accounts for fractional atomic existence. This allows us to calculate the gradient of the Gibbs free energy with respect to both the Cartesian coordinates of atoms and their existence. Using this we propose a gradient-based grand canonical optimization method and document its capabilities for a Cu(110) surface oxide.<\/jats:p>","DOI":"10.1088\/2632-2153\/ae1bf6","type":"journal-article","created":{"date-parts":[[2025,11,5]],"date-time":"2025-11-05T22:53:13Z","timestamp":1762383193000},"page":"045049","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Gradient-based grand canonical optimization enabled by graph neural networks with fractional atomic existence"],"prefix":"10.1088","volume":"6","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-3550-8379","authenticated-orcid":true,"given":"Mads-Peter Verner","family":"Christiansen","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7849-6347","authenticated-orcid":true,"given":"Bj\u00f8rk","family":"Hammer","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"266","published-online":{"date-parts":[[2025,11,18]]},"reference":[{"key":"mlstae1bf6bib1","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.98.146401","type":"journal-article","article-title":"Generalized neural-network representation of high-dimensional potential-energy surfaces","volume":"98","author":"Behler","year":"2007","journal-title":"Phys. Rev. Lett."},{"key":"mlstae1bf6bib2","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.104.136403","type":"journal-article","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."},{"article-title":"Schnet: a continuous-filter convolutional neural network for modeling quantum interactions","year":"2017","author":"Sch\u00fctt","key":"mlstae1bf6bib3","type":"conference-proceedings"},{"article-title":"Neural message passing for quantum chemistry","year":"2017","author":"Gilmer","key":"mlstae1bf6bib4","type":"conference-proceedings"},{"key":"mlstae1bf6bib5","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.120.145301","type":"journal-article","article-title":"Crystal graph convolutional neural networks for an accurate and interpretable prediction of material properties","volume":"120","author":"Xie","year":"2018","journal-title":"Phys. Rev. Lett."},{"key":"mlstae1bf6bib6","doi-asserted-by":"publisher","first-page":"3678","DOI":"10.1021\/acs.jctc.9b00181","type":"journal-article","article-title":"PhysNet: a neural network for predicting energies, forces, dipole moments and partial charges","volume":"15","author":"Unke","year":"2019","journal-title":"J. Chem. Theory Comput."},{"article-title":"Tensor field networks: rotation- and translation-equivariant neural networks for 3D point clouds","year":"2018","author":"Thomas","key":"mlstae1bf6bib7","type":"preprint"},{"article-title":"Directional message passing for molecular graphs","year":"2019","author":"Gasteiger","key":"mlstae1bf6bib8","type":"conference-proceedings"},{"article-title":"Cormorant: covariant molecular neural networks","year":"2019","author":"Anderson","key":"mlstae1bf6bib9","type":"preprint"},{"key":"mlstae1bf6bib10","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":"mlstae1bf6bib11","doi-asserted-by":"publisher","first-page":"2453","DOI":"10.1038\/s41467-022-29939-5","type":"journal-article","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":"mlstae1bf6bib12","first-page":"pp 11423","type":"conference-proceedings","article-title":"Mace: Higher order equivariant message passing neural networks for fast and accurate force fields","author":"Batatia","year":"2022"},{"key":"mlstae1bf6bib13","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/sdata.2018.62","type":"journal-article","article-title":"The alexandria library, a quantum-chemical database of molecular properties for force field development","volume":"5","author":"Ghahremanpour","year":"2018","journal-title":"Sci. Data"},{"key":"mlstae1bf6bib14","doi-asserted-by":"publisher","first-page":"246","DOI":"10.1038\/s41565-017-0035-5","type":"journal-article","article-title":"Two-dimensional materials from high-throughput computational exfoliation of experimentally known compounds","volume":"13","author":"Mounet","year":"2018","journal-title":"Nat. Nanotechnol."},{"key":"mlstae1bf6bib15","doi-asserted-by":"publisher","first-page":"1031","DOI":"10.1038\/s42256-023-00716-3","type":"journal-article","article-title":"CHGNet as a pretrained universal neural network potential for charge-informed atomistic modelling","volume":"5","author":"Deng","year":"2023","journal-title":"Nat. Mach. Intell."},{"article-title":"The OpenLAM challenges","year":"2025","author":"Peng","key":"mlstae1bf6bib16","type":"preprint"},{"key":"mlstae1bf6bib17","doi-asserted-by":"publisher","first-page":"308","DOI":"10.1038\/s41597-023-02200-4","type":"journal-article","article-title":"A database of low-energy atomically precise nanoclusters","volume":"10","author":"Manna","year":"2023","journal-title":"Sci. Data"},{"article-title":"Open materials 2024 (OMat24) inorganic materials dataset and models","year":"2024","author":"Barroso-Luque","key":"mlstae1bf6bib18","type":"preprint"},{"key":"mlstae1bf6bib19","doi-asserted-by":"publisher","first-page":"80","DOI":"10.1038\/s41586-023-06735-9","type":"journal-article","article-title":"Scaling deep learning for materials discovery","volume":"624","author":"Merchant","year":"2023","journal-title":"Nature"},{"article-title":"A foundation model for atomistic materials chemistry","year":"2024","author":"Batatia","key":"mlstae1bf6bib20","type":"preprint"},{"key":"mlstae1bf6bib21","doi-asserted-by":"publisher","first-page":"4857","DOI":"10.1021\/acs.jctc.4c00190","type":"journal-article","article-title":"Scalable parallel algorithm for graph neural network interatomic potentials in molecular dynamics simulations","volume":"20","author":"Park","year":"2024","journal-title":"J. Chem. Theory Comput."},{"article-title":"ORB: a fast, scalable neural network potential","year":"2024","author":"Neumann","key":"mlstae1bf6bib22","type":"preprint"},{"article-title":"Data-efficient multi-fidelity training for high-fidelity machine learning interatomic potentials","year":"2024","author":"Kim","key":"mlstae1bf6bib23","type":"preprint"},{"article-title":"MatterSim: a deep learning atomistic model across elements, temperatures and pressures","year":"2024","author":"Yang","key":"mlstae1bf6bib24","type":"preprint"},{"key":"mlstae1bf6bib25","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevX.14.021036","type":"journal-article","article-title":"Graph atomic cluster expansion for semilocal interactions beyond equivariant message passing","volume":"14","author":"Bochkarev","year":"2024","journal-title":"Phys. Rev. X"},{"article-title":"Learning smooth and expressive interatomic potentials for physical property prediction","year":"2025","author":"Fu","key":"mlstae1bf6bib26","type":"preprint"},{"key":"mlstae1bf6bib27","doi-asserted-by":"publisher","first-page":"196","DOI":"10.1038\/s41524-025-01691-6","type":"journal-article","article-title":"Machine learning and data-driven methods in computational surface and interface science","volume":"11","author":"H\u00f6rmann","year":"2025","journal-title":"npj Comput. Mater."},{"key":"mlstae1bf6bib28","doi-asserted-by":"publisher","first-page":"671","DOI":"10.1126\/science.220.4598.671","type":"journal-article","article-title":"Optimization by simulated annealing","volume":"220","author":"Kirkpatrick","year":"1983","journal-title":"Science"},{"key":"mlstae1bf6bib29","doi-asserted-by":"publisher","first-page":"5111","DOI":"10.1021\/jp970984n","type":"journal-article","article-title":"Global optimization by basin-hopping and the lowest energy structures of Lennard-Jones clusters containing up to 110 atoms","volume":"101","author":"Wales","year":"1997","journal-title":"J. Phys. Chem. A"},{"key":"mlstae1bf6bib30","first-page":"pp 385","type":"book","article-title":"Nudged elastic band method for finding minimum energy paths of transitions","author":"J\u00f3nsson","year":"1998"},{"key":"mlstae1bf6bib31","doi-asserted-by":"publisher","DOI":"10.1088\/0953-8984\/23\/5\/053201","type":"journal-article","article-title":"Ab initio random structure searching","volume":"23","author":"Pickard","year":"2011","journal-title":"J. Phys.: Condens. Matter"},{"key":"mlstae1bf6bib32","doi-asserted-by":"publisher","first-page":"4193","DOI":"10.1039\/b305686d","type":"journal-article","article-title":"Evolving better nanoparticles: genetic algorithms for optimising cluster geometries","volume":"0","author":"Johnston","year":"2003","journal-title":"Dalton Trans."},{"key":"mlstae1bf6bib33","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevB.97.195424","type":"journal-article","article-title":"Neural-network-enhanced evolutionary algorithm applied to supported metal nanoparticles","volume":"97","author":"Kolsbjerg","year":"2018","journal-title":"Phys. Rev. B"},{"key":"mlstae1bf6bib34","doi-asserted-by":"publisher","DOI":"10.1063\/5.0014876","type":"journal-article","article-title":"Global optimization of copper clusters at the zno(101 \u00af0) surface using a dft-based neural network potential and genetic algorithms","volume":"153","author":"Leandro Paleico","year":"2020","journal-title":"J. Chem. Phys."},{"key":"mlstae1bf6bib35","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41524-021-00537-1","type":"journal-article","article-title":"Evolutionary computing and machine learning for discovering of low-energy defect configurations","volume":"7","author":"Arrigoni","year":"2021","journal-title":"npj Comput. Mater."},{"key":"mlstae1bf6bib36","doi-asserted-by":"publisher","first-page":"1601","DOI":"10.1021\/acs.jpcc.0c09531","type":"journal-article","article-title":"The XtalOpt evolutionary algorithm for crystal structure prediction","volume":"125","author":"Falls","year":"2021","journal-title":"J. Phys. Chem. C"},{"key":"mlstae1bf6bib37","doi-asserted-by":"publisher","first-page":"nwad128","DOI":"10.1093\/nsr\/nwad128","type":"journal-article","article-title":"MAGUS: machine learning and graph theory assisted universal structure searcher","volume":"10","author":"Wang","year":"2023","journal-title":"Nat. Sci. Rev."},{"key":"mlstae1bf6bib38","doi-asserted-by":"publisher","DOI":"10.1016\/j.cpc.2023.109065","type":"journal-article","article-title":"Clinamen2: functional-style evolutionary optimization in Python for atomistic structure searches","volume":"297","author":"Wanzenb\u00f6ck","year":"2024","journal-title":"Comput. Phys. Commun."},{"key":"mlstae1bf6bib39","first-page":"pp 1470","type":"conference-proceedings","article-title":"Parallel and distributed thompson sampling for large-scale accelerated exploration of chemical space","author":"Miguel Hern\u00e1ndez-Lobato","year":"2017"},{"key":"mlstae1bf6bib40","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevMaterials.2.013803","type":"journal-article","article-title":"Crystal structure prediction accelerated by Bayesian optimization","volume":"2","author":"Yamashita","year":"2018","journal-title":"Phys. Rev. Mater."},{"key":"mlstae1bf6bib41","doi-asserted-by":"publisher","first-page":"35","DOI":"10.1038\/s41524-019-0175-2","type":"journal-article","article-title":"Bayesian inference of atomistic structure in functional materials","volume":"5","author":"Todorovi\u0107","year":"2019","journal-title":"npj Comput. Mater."},{"key":"mlstae1bf6bib42","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.124.086102","type":"journal-article","article-title":"Efficient global structure optimization with a machine-learned surrogate model","volume":"124","author":"Bisbo","year":"2020","journal-title":"Phys. Rev. Lett."},{"key":"mlstae1bf6bib43","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevB.103.174114","type":"journal-article","article-title":"Global optimization of atomic structures with gradient-enhanced Gaussian process regression","volume":"103","author":"Kaappa","year":"2021","journal-title":"Phys. Rev. B"},{"key":"mlstae1bf6bib44","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevMaterials.8.123802","type":"journal-article","article-title":"Bayesian optimization of atomic structures with prior probabilities from universal interatomic potentials","volume":"8","author":"Lyngby","year":"2024","journal-title":"Phys. Rev. Mater."},{"key":"mlstae1bf6bib45","doi-asserted-by":"publisher","first-page":"1177","DOI":"10.1002\/jcc.20621","type":"journal-article","article-title":"Global minimum structure searches via particle swarm optimization","volume":"28","author":"Call","year":"2007","journal-title":"J. Comput. Chem."},{"key":"mlstae1bf6bib46","doi-asserted-by":"publisher","DOI":"10.1063\/1.4746757","type":"journal-article","article-title":"Particle-swarm structure prediction on clusters","volume":"137","author":"Lv","year":"2012","journal-title":"J. Chem. Phys."},{"key":"mlstae1bf6bib47","doi-asserted-by":"publisher","first-page":"406","DOI":"10.1016\/j.commatsci.2015.09.037","type":"journal-article","article-title":"CALYPSO structure prediction method and its wide application","volume":"112","author":"Wang","year":"2016","journal-title":"Comput. Mater. Sci."},{"article-title":"Symmetry-aware actor-critic for 3D molecular design","year":"2020","author":"Simm","key":"mlstae1bf6bib48","type":"preprint"},{"key":"mlstae1bf6bib49","doi-asserted-by":"publisher","DOI":"10.1088\/2632-2153\/ac3eb4","type":"journal-article","article-title":"Generating stable molecules using imitation and reinforcement learning","volume":"3","author":"Ager Meldgaard","year":"2021","journal-title":"Mach. Learn.: Sci. Technol."},{"key":"mlstae1bf6bib50","doi-asserted-by":"publisher","first-page":"3169","DOI":"10.1021\/acs.jcim.2c00373","type":"journal-article","article-title":"Exploring potential energy surfaces using reinforcement machine learning","volume":"62","author":"Mills","year":"2022","journal-title":"J. Chem. Inf. Model."},{"key":"mlstae1bf6bib51","doi-asserted-by":"publisher","DOI":"10.1063\/5.0201421","type":"journal-article","article-title":"Cascading symmetry constraint during machine learning-enabled structural search for sulfur-induced Cu(111)-(43\u00d743) surface reconstruction","volume":"160","author":"Brix","year":"2024","journal-title":"J. Chem. Phys."},{"key":"mlstae1bf6bib52","doi-asserted-by":"publisher","DOI":"10.1063\/5.0207801","type":"journal-article","article-title":"Accelerating structure search using atomistic graph-based classifiers","volume":"161","author":"M\u00f8ller Slavensky","year":"2024","journal-title":"J. Chem. Phys."},{"key":"mlstae1bf6bib53","doi-asserted-by":"crossref","DOI":"10.1038\/s41524-025-01749-5","type":"preprint","article-title":"RAFFLE: active learning accelerated interface structure prediction","author":"Taylor","year":"2025"},{"key":"mlstae1bf6bib54","doi-asserted-by":"publisher","first-page":"33","DOI":"10.1007\/s11244-013-0160-9","type":"journal-article","article-title":"Thomas Bligaard and Tejs Vegge. Genetic algorithm procreation operators for alloy nanoparticle catalysts","volume":"57","author":"Lysgaard","year":"2014","journal-title":"Top. Catal."},{"key":"mlstae1bf6bib55","doi-asserted-by":"publisher","first-page":"61","DOI":"10.1039\/D4FD00134F","type":"journal-article","article-title":"Beyond theory-driven discovery: introducing hot random search and datum-derived structures","volume":"256","author":"Pickard","year":"2025","journal-title":"Faraday Discuss."},{"key":"mlstae1bf6bib56","doi-asserted-by":"publisher","first-page":"14817","DOI":"10.1039\/C5NR03903G","type":"journal-article","article-title":"Global minimization of gold clusters by combining neural network potentials and the basin-hopping method","volume":"7","author":"Ouyang","year":"2015","journal-title":"Nanoscale"},{"key":"mlstae1bf6bib57","doi-asserted-by":"publisher","first-page":"31","DOI":"10.1039\/C8FD00055G","type":"journal-article","article-title":"Accelerating CALYPSO structure prediction by data-driven learning of a potential energy surface","volume":"211","author":"Tong","year":"2018","journal-title":"Faraday Discuss."},{"key":"mlstae1bf6bib58","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.125.206101","type":"journal-article","article-title":"IrO2 surface complexions identified through machine learning and surface investigations","volume":"125","author":"Timmermann","year":"2020","journal-title":"Phys. Rev. Lett."},{"key":"mlstae1bf6bib59","doi-asserted-by":"publisher","DOI":"10.1063\/5.0094165","type":"journal-article","article-title":"Atomistic global optimization X: a Python package for optimization of atomistic structures","volume":"157","author":"Christiansen","year":"2022","journal-title":"J. Chem. Phys."},{"key":"mlstae1bf6bib60","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41524-023-01104-6","type":"journal-article","article-title":"Hyperactive learning for data-driven interatomic potentials","volume":"9","author":"van der Oord","year":"2023","journal-title":"npj Comput. Mater."},{"key":"mlstae1bf6bib61","doi-asserted-by":"publisher","first-page":"230","DOI":"10.1038\/s43588-023-00406-5","type":"journal-article","article-title":"Uncertainty-driven dynamics for active learning of interatomic potentials","volume":"3","author":"Kulichenko","year":"2023","journal-title":"Nat. Comput. Sci."},{"key":"mlstae1bf6bib62","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41524-024-01254-1","type":"journal-article","article-title":"Uncertainty-biased molecular dynamics for learning uniformly accurate interatomic potentials","volume":"10","author":"Zaverkin","year":"2024","journal-title":"npj Comput. Mater."},{"key":"mlstae1bf6bib63","doi-asserted-by":"publisher","DOI":"10.1063\/5.0156218","type":"journal-article","article-title":"Generating candidates in global optimization algorithms using complementary energy landscapes","volume":"159","author":"M\u00f8ller Slavensky","year":"2023","journal-title":"J. Chem. Phys."},{"key":"mlstae1bf6bib64","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevResearch.5.013189","type":"journal-article","article-title":"Targeting high symmetry in structure predictions by biasing the potential energy surface","volume":"5","author":"Huber","year":"2023","journal-title":"Phys. Rev. Res."},{"key":"mlstae1bf6bib65","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevB.99.054102","type":"journal-article","article-title":"Hyperspatial optimization of structures","volume":"99","author":"Pickard","year":"2019","journal-title":"Phys. Rev. B"},{"key":"mlstae1bf6bib66","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.127.166001","type":"journal-article","article-title":"Atomic structure optimization with machine-learning enabled interpolation between chemical elements","volume":"127","author":"Kaappa","year":"2021","journal-title":"Phys. Rev. Lett."},{"key":"mlstae1bf6bib67","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevB.107.214101","type":"journal-article","article-title":"Machine-learning-enabled optimization of atomic structures using atoms with fractional existence","volume":"107","author":"Larsen","year":"2023","journal-title":"Phys. Rev. B"},{"key":"mlstae1bf6bib68","doi-asserted-by":"publisher","first-page":"222","DOI":"10.1038\/s41524-025-01656-9","type":"journal-article","article-title":"Global atomic structure optimization through machine-learning-enabled barrier circumvention in extra dimensions","volume":"11","author":"Larsen","year":"2025","journal-title":"npj Comput. Mater."},{"key":"mlstae1bf6bib69","doi-asserted-by":"publisher","DOI":"10.1063\/5.0084545","type":"journal-article","article-title":"Automated search for optimal surface phases (ASOPs) in grand canonical ensemble powered by machine learning","volume":"156","author":"Chen","year":"2022","journal-title":"J. Chem. Phys."},{"key":"mlstae1bf6bib70","doi-asserted-by":"publisher","first-page":"1162","DOI":"10.1021\/jacsau.3c00038","type":"journal-article","article-title":"In situ structure of a Mo-Doped Pt\u2013Ni catalyst during electrochemical oxygen reduction resolved from machine learning-based grand canonical global optimization","volume":"3","author":"Li","year":"2023","journal-title":"JACS Au"},{"key":"mlstae1bf6bib71","doi-asserted-by":"publisher","DOI":"10.7566\/JPSJ.94.044802","type":"journal-article","article-title":"Machine learning supported annealing for prediction of grand canonical crystal structures","volume":"94","author":"Couzini\u00e9","year":"2025","journal-title":"J. Phys. Soc. Japan"},{"key":"mlstae1bf6bib72","doi-asserted-by":"publisher","first-page":"696","DOI":"10.1039\/D4CP03801K","type":"journal-article","article-title":"GOCIA: a grand canonical global optimizer for clusters, interfaces and adsorbates","volume":"27","author":"Zhang","year":"2025","journal-title":"Phys. Chem. Chem. Phys."},{"key":"mlstae1bf6bib73","doi-asserted-by":"publisher","first-page":"1451","DOI":"10.1021\/ct7000039","type":"journal-article","article-title":"Continuous fractional component Monte Carlo: an adaptive biasing method for open system atomistic simulations","volume":"3","author":"Shi","year":"2007","journal-title":"J. Chem. Theory Comput."},{"key":"mlstae1bf6bib74","doi-asserted-by":"publisher","first-page":"942","DOI":"10.1021\/ct4009766","type":"journal-article","article-title":"A comparison of advanced Monte Carlo methods for open systems: CFCMC vs CBMC","volume":"10","author":"Torres-Knoop","year":"2014","journal-title":"J. Chem. Theory Comput."},{"key":"mlstae1bf6bib75","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.97.170201","type":"journal-article","article-title":"Structural relaxation made simple","volume":"97","author":"Bitzek","year":"2006","journal-title":"Phys. Rev. Lett."},{"key":"mlstae1bf6bib76","doi-asserted-by":"publisher","DOI":"10.1063\/5.0268264","type":"journal-article","article-title":"\u0394-model correction of foundation model based on the model\u2019s own understanding","volume":"162","author":"Christiansen","year":"2025","journal-title":"J. Chem. Phys."},{"key":"mlstae1bf6bib77","type":"journal-article","article-title":"Gradient-based grand canonical optimization enabled by graph neural networks with fractional atomic existence","author":"Mads-Peter","year":"2025","journal-title":"Zenodo"},{"key":"mlstae1bf6bib78","first-page":"pp 8026","type":"conference-proceedings","article-title":"PyTorch: an imperative style, high-performance deep learning library","author":"Paszke","year":"2019"},{"article-title":"Fast graph representation learning with PyTorch geometric","year":"2019","author":"Fey","key":"mlstae1bf6bib79","type":"preprint"},{"key":"mlstae1bf6bib80","doi-asserted-by":"publisher","DOI":"10.5281\/zenodo.15706795)","type":"other","article-title":"PyTorch lightning","author":"(The PyTorch Lightning team)","year":"2025"},{"key":"mlstae1bf6bib81","doi-asserted-by":"publisher","first-page":"1691","DOI":"10.1103\/PhysRevLett.49.1691","type":"journal-article","article-title":"Density-functional theory for fractional particle number: derivative discontinuities of the energy","volume":"49","author":"Perdew","year":"1982","journal-title":"Phys. Rev. Lett."},{"key":"mlstae1bf6bib82","doi-asserted-by":"publisher","first-page":"2027","DOI":"10.1103\/PhysRevLett.65.2027","type":"journal-article","article-title":"Oxygen chemisorption on Cu(110): a model for thephc(6\u00d72) structure","volume":"65","author":"Feidenhans\u2019l","year":"1990","journal-title":"Phys. Rev. Lett."},{"key":"mlstae1bf6bib83","doi-asserted-by":"publisher","DOI":"10.1088\/1361-648X\/aa680e","type":"journal-article","article-title":"The atomic simulation environment\u2014a Python library for working with atoms","volume":"29","author":"Larsen","year":"2017","journal-title":"J. Phys.: Condens. Matter"},{"key":"mlstae1bf6bib84","doi-asserted-by":"publisher","DOI":"10.1063\/5.0182685","type":"journal-article","article-title":"GPAW: an open Python package for electronic structure calculations","volume":"160","author":"J\u00f8rgen Mortensen","year":"2024","journal-title":"J. Chem. Phys."},{"article-title":"Crystal diffusion variational autoencoder for periodic material generation","year":"2022","author":"Xie","key":"mlstae1bf6bib85","type":"preprint"},{"article-title":"Equivariant diffusion for molecule generation in 3D","year":"2022","author":"Hoogeboom","key":"mlstae1bf6bib86","type":"preprint"},{"key":"mlstae1bf6bib87","doi-asserted-by":"publisher","first-page":"232","DOI":"10.1038\/s41524-022-00923-3","type":"journal-article","article-title":"Data-driven discovery of 2D materials by deep generative models","volume":"8","author":"Lyngby","year":"2022","journal-title":"npj Comput. Mater."},{"key":"mlstae1bf6bib88","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevB.110.235427","type":"journal-article","article-title":"Generative diffusion model for surface structure discovery","volume":"110","author":"R\u00f8nne","year":"2024","journal-title":"Phys. Rev. B"}],"container-title":["Machine Learning: Science and Technology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae1bf6","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae1bf6\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae1bf6","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae1bf6\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae1bf6\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae1bf6\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae1bf6\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae1bf6\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,18]],"date-time":"2025-11-18T09:40:20Z","timestamp":1763458820000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae1bf6"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,11,18]]},"references-count":88,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2025,11,18]]},"published-print":{"date-parts":[[2025,12,30]]}},"URL":"https:\/\/doi.org\/10.1088\/2632-2153\/ae1bf6","relation":{},"ISSN":["2632-2153"],"issn-type":[{"type":"electronic","value":"2632-2153"}],"subject":[],"published":{"date-parts":[[2025,11,18]]},"assertion":[{"value":"Gradient-based grand canonical optimization enabled by graph neural networks with fractional atomic existence","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":"2025-07-25","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-11-05","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-11-18","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}