{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,23]],"date-time":"2026-04-23T19:17:17Z","timestamp":1776971837990,"version":"3.51.4"},"reference-count":53,"publisher":"IOP Publishing","issue":"4","license":[{"start":{"date-parts":[[2022,12,29]],"date-time":"2022-12-29T00:00:00Z","timestamp":1672272000000},"content-version":"vor","delay-in-days":28,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2022,12,29]],"date-time":"2022-12-29T00:00:00Z","timestamp":1672272000000},"content-version":"tdm","delay-in-days":28,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"DOI":"10.13039\/501100002241","name":"Japan Science and Technology Agency","doi-asserted-by":"crossref","award":["JPMJMI19G1"],"award-info":[{"award-number":["JPMJMI19G1"]}],"id":[{"id":"10.13039\/501100002241","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":[[2022,12,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Material development involves laborious processes to explore the vast materials space. The key to accelerating these processes is understanding the structure-functionality relationships of materials. Machine learning has enabled large-scale analysis of underlying relationships between materials via their vector representations, or embeddings. However, the learning of material embeddings spanning most known inorganic materials has remained largely unexplored due to the expert knowledge and efforts required to annotate large-scale materials data. Here we show that our self-supervised deep learning approach can successfully learn material embeddings from crystal structures of over 120\u2009000 materials, without any annotations, to capture the structure-functionality relationships among materials. These embeddings revealed the profound similarity between materials, or \u2018materials concepts\u2019, such as cuprate superconductors and lithium-ion battery materials from the unannotated structural data. Consequently, our results enable us to both draw a large-scale map of the materials space, capturing various materials concepts, and measure the functionality-aware similarities between materials. Our findings will enable more strategic approaches to material development.<\/jats:p>","DOI":"10.1088\/2632-2153\/aca23d","type":"journal-article","created":{"date-parts":[[2022,11,11]],"date-time":"2022-11-11T22:42:40Z","timestamp":1668206560000},"page":"045034","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":24,"title":["Self-supervised learning of materials concepts from crystal structures via deep neural networks"],"prefix":"10.1088","volume":"3","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-0019-4832","authenticated-orcid":true,"given":"Yuta","family":"Suzuki","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3361-4861","authenticated-orcid":false,"given":"Tatsunori","family":"Taniai","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1281-8099","authenticated-orcid":false,"given":"Kotaro","family":"Saito","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9014-1389","authenticated-orcid":false,"given":"Yoshitaka","family":"Ushiku","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3285-9093","authenticated-orcid":true,"given":"Kanta","family":"Ono","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"266","published-online":{"date-parts":[[2022,12,29]]},"reference":[{"key":"mlstaca23dbib1","author":"De Graef","year":"2012"},{"key":"mlstaca23dbib2","author":"Callister","year":"2007"},{"key":"mlstaca23dbib3","author":"Anderson","year":"1997"},{"key":"mlstaca23dbib4","author":"Coey","year":"2010"},{"key":"mlstaca23dbib5","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41467-020-15355-0","article-title":"A reflection on lithium-ion battery cathode chemistry","volume":"11","author":"Manthiram","year":"2020","journal-title":"Nat. Commun."},{"key":"mlstaca23dbib6","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":"mlstaca23dbib7","author":"Lookman","year":"2015"},{"key":"mlstaca23dbib8","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":"mlstaca23dbib9","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-017-17299-w","article-title":"A universal 3D voxel descriptor for solid-state material informatics with deep convolutional neural networks","volume":"7","author":"Kajita","year":"2017","journal-title":"Sci. Rep."},{"key":"mlstaca23dbib10","doi-asserted-by":"publisher","DOI":"10.1063\/1.5019779","article-title":"SchNet\u2014a deep learning architecture for molecules and materials","volume":"148","author":"Sch\u00fctt","year":"2018","journal-title":"J. Chem. Phys."},{"key":"mlstaca23dbib11","doi-asserted-by":"publisher","first-page":"2775","DOI":"10.1038\/s41467-018-05169-6","article-title":"Insightful classification of crystal structures using deep learning","volume":"9","author":"Ziletti","year":"2018","journal-title":"Nat. Commun."},{"key":"mlstaca23dbib12","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.120.145301","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":"mlstaca23dbib13","doi-asserted-by":"publisher","first-page":"3564","DOI":"10.1021\/acs.chemmater.9b01294","article-title":"Graph networks as a universal machine learning framework for molecules and crystals","volume":"31","author":"Chen","year":"2019","journal-title":"Chem. Mater."},{"key":"mlstaca23dbib14","doi-asserted-by":"publisher","first-page":"7503","DOI":"10.1039\/C9SC02097G","article-title":"A generalized deep learning approach for local structure identification in molecular simulations","volume":"10","author":"DeFever","year":"2019","journal-title":"Chem. Sci."},{"key":"mlstaca23dbib15","article-title":"Fast and uncertainty-aware directional message passing for non-equilibrium molecules","author":"Gasteiger","year":"2020"},{"key":"mlstaca23dbib16","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41524-021-00650-1","article-title":"Atomistic Line Graph Neural Network for improved materials property predictions","volume":"7","author":"Choudhary","year":"2021","journal-title":"npj Comput. Mater."},{"key":"mlstaca23dbib17","doi-asserted-by":"publisher","first-page":"718","DOI":"10.1038\/s43588-022-00349-3","article-title":"A universal graph deep learning interatomic potential for the periodic table","volume":"2","author":"Chen","year":"2022","journal-title":"Nat. Comput. Sci."},{"key":"mlstaca23dbib18","doi-asserted-by":"publisher","DOI":"10.1016\/j.patter.2022.100491","article-title":"Scalable deeper graph neural networks for high-performance materials property prediction","volume":"3","author":"Omee","year":"2022","journal-title":"Patter"},{"key":"mlstaca23dbib19","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":"mlstaca23dbib20","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":"mlstaca23dbib21","doi-asserted-by":"publisher","first-page":"2326","DOI":"10.1021\/acs.jpclett.5b00831","article-title":"Machine learning predictions of molecular properties: accurate many-body potentials and nonlocality in chemical space","volume":"6","author":"Hansen","year":"2015","journal-title":"J. Phys. Chem. Lett."},{"key":"mlstaca23dbib22","doi-asserted-by":"publisher","first-page":"1094","DOI":"10.1002\/qua.24917","article-title":"Crystal structure representations for machine learning models of formation energies","volume":"115","author":"Faber","year":"2015","journal-title":"Int. J. Quantum Chem."},{"key":"mlstaca23dbib23","doi-asserted-by":"publisher","first-page":"735","DOI":"10.1021\/cm503507h","article-title":"Materials cartography: representing and mining materials space using structural and electronic fingerprints","volume":"27","author":"Isayev","year":"2015","journal-title":"Chem. Mater."},{"key":"mlstaca23dbib24","doi-asserted-by":"publisher","DOI":"10.1038\/ncomms15679","article-title":"Universal fragment descriptors for predicting properties of inorganic crystals","volume":"8","author":"Isayev","year":"2017","journal-title":"Nat. Commun."},{"key":"mlstaca23dbib25","doi-asserted-by":"publisher","first-page":"E6411","DOI":"10.1073\/pnas.1801181115","article-title":"Learning atoms for materials discovery","volume":"115","author":"Zhou","year":"2018","journal-title":"Proc. Natl Acad. Sci. USA"},{"key":"mlstaca23dbib26","doi-asserted-by":"publisher","first-page":"95","DOI":"10.1038\/s41586-019-1335-8","article-title":"Unsupervised word embeddings capture latent knowledge from materials science literature","volume":"571","author":"Tshitoyan","year":"2019","journal-title":"Nature"},{"key":"mlstaca23dbib27","doi-asserted-by":"publisher","first-page":"158","DOI":"10.1021\/jacs.8b03913","article-title":"Crystal structure prediction via deep learning","volume":"140","author":"Ryan","year":"2018","journal-title":"J. Am. Chem. Soc."},{"key":"mlstaca23dbib28","doi-asserted-by":"publisher","DOI":"10.1063\/1.5047803","article-title":"Hierarchical visualization of materials space with graph convolutional neural networks","volume":"149","author":"Xie","year":"2018","journal-title":"J. Chem. Phys."},{"key":"mlstaca23dbib29","doi-asserted-by":"publisher","first-page":"144","DOI":"10.1038\/s42256-020-00284-w","article-title":"Mapping the space of chemical reactions using attention-based neural networks","volume":"3","author":"Schwaller","year":"2021","journal-title":"Nat. Mach. Intell."},{"key":"mlstaca23dbib30","doi-asserted-by":"publisher","first-page":"433","DOI":"10.1016\/j.matt.2020.04.016","article-title":"Crystal site feature embedding enables exploration of large chemical spaces","volume":"3","author":"Choubisa","year":"2020","journal-title":"Matter"},{"key":"mlstaca23dbib31","article-title":"Data-driven approach to encoding and decoding 3-D crystal structures","author":"Hoffmann","year":"2019"},{"key":"mlstaca23dbib32","doi-asserted-by":"publisher","first-page":"1370","DOI":"10.1016\/j.matt.2019.08.017","article-title":"Inverse design of solid-state materials via a continuous representation","volume":"1","author":"Noh","year":"2019","journal-title":"Matter"},{"key":"mlstaca23dbib33","doi-asserted-by":"publisher","first-page":"4871","DOI":"10.1039\/D0SC00594K","article-title":"Machine-enabled inverse design of inorganic solid materials: promises and challenges","volume":"11","author":"Noh","year":"2020","journal-title":"Chem. Sci."},{"key":"mlstaca23dbib34","doi-asserted-by":"publisher","first-page":"4518","DOI":"10.1021\/acs.jcim.0c00464","article-title":"3-D inorganic crystal structure generation and property prediction via representation learning","volume":"60","author":"Court","year":"2020","journal-title":"J. Chem. Inf. Model."},{"key":"mlstaca23dbib35","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41524-021-00526-4","article-title":"Constrained crystals deep convolutional generative adversarial network for the inverse design of crystal structures","volume":"7","author":"Long","year":"2021","journal-title":"npj Comput. Mater."},{"key":"mlstaca23dbib36","doi-asserted-by":"publisher","first-page":"314","DOI":"10.1016\/j.matt.2021.11.032","article-title":"An invertible crystallographic representation for general inverse design of inorganic crystals with targeted properties","volume":"5","author":"Ren","year":"2022","journal-title":"Matter"},{"key":"mlstaca23dbib37","first-page":"pp 2051","article-title":"Multi-task self-supervised visual learning","author":"Doersch","year":"2017"},{"key":"mlstaca23dbib38","doi-asserted-by":"publisher","first-page":"1066","DOI":"10.3390\/sym11091066","article-title":"Deep metric learning: a survey","volume":"11","author":"Kaya","year":"2019","journal-title":"Symmetry"},{"key":"mlstaca23dbib39","doi-asserted-by":"publisher","DOI":"10.1038\/natrevmats.2017.33","article-title":"2D transition metal dichalcogenides","volume":"2","author":"Manzeli","year":"2017","journal-title":"Nat. Rev. Mater."},{"key":"mlstaca23dbib40","doi-asserted-by":"publisher","first-page":"2388","DOI":"10.1143\/JJAP.29.2388","article-title":"New classification method for layered copper oxide compounds and its application to design of new high T c superconductors","volume":"29","author":"Tokura","year":"1990","journal-title":"Jpn. J. Appl. Phys."},{"key":"mlstaca23dbib41","doi-asserted-by":"publisher","first-page":"47","DOI":"10.1038\/s41586-018-0631-z","article-title":"Magnetism in two-dimensional van der Waals materials","volume":"563","author":"Burch","year":"2018","journal-title":"Nature"},{"key":"mlstaca23dbib42","doi-asserted-by":"publisher","first-page":"56","DOI":"10.1038\/363056a0","article-title":"Superconductivity above 130 K in the Hg\u2013Ba\u2013Ca\u2013Cu\u2013O system","volume":"363","author":"Schilling","year":"1993","journal-title":"Nature"},{"key":"mlstaca23dbib43","doi-asserted-by":"publisher","first-page":"510","DOI":"10.1038\/334510a0","article-title":"A new high-T c TlBa2Ca3Cu4O11 superconductor with T c> 120K","volume":"334","author":"Ihara","year":"1988","journal-title":"Nature"},{"key":"mlstaca23dbib44","first-page":"pp 689","article-title":"Multimodal deep learning","author":"Ngiam","year":"2011"},{"key":"mlstaca23dbib45","article-title":"BERT: pre-training of deep bidirectional transformers for language understanding","author":"Devlin","year":"2018"},{"key":"mlstaca23dbib46","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41570-018-0121","article-title":"Inverse design in search of materials with target functionalities","volume":"2","author":"Zunger","year":"2018","journal-title":"Nat. Rev. Chem."},{"key":"mlstaca23dbib47","doi-asserted-by":"publisher","first-page":"1272","DOI":"10.1107\/S0021889811038970","article-title":"VESTA 3 for three-dimensional visualization of crystal, volumetric and morphology data","volume":"44","author":"Momma","year":"2011","journal-title":"J. Appl. Crystallogr."},{"key":"mlstaca23dbib48","doi-asserted-by":"publisher","first-page":"314","DOI":"10.1016\/j.commatsci.2012.10.028","article-title":"Python Materials Genomics (pymatgen): a robust, open-source python library for materials analysis","volume":"68","author":"Ong","year":"2013","journal-title":"Comput. Mater. Sci."},{"key":"mlstaca23dbib49","doi-asserted-by":"publisher","first-page":"486","DOI":"10.1107\/S205225251700714X","article-title":"Classification of crystal structure using a convolutional neural network","volume":"4","author":"Park","year":"2017","journal-title":"IUCrJ"},{"key":"mlstaca23dbib50","first-page":"pp 448","article-title":"Batch normalization: accelerating deep network training by reducing internal covariate shift","author":"Ioffe","year":"2015"},{"key":"mlstaca23dbib51","doi-asserted-by":"publisher","first-page":"3984","DOI":"10.1109\/CVPR.2017.424","article-title":"Online asymmetric similarity learning for cross-modal retrieval","author":"Wu","year":"2017"},{"key":"mlstaca23dbib52","article-title":"Adam: a method for stochastic optimization","author":"Kingma","year":"2015"},{"key":"mlstaca23dbib53","doi-asserted-by":"publisher","DOI":"10.6084\/m9.figshare.21717824","article-title":"Self-supervised learning of materials concepts from crystal structures via deep neural networks","author":"Suzuki","year":"2022","journal-title":"figshare"}],"container-title":["Machine Learning: Science and Technology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/aca23d","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/aca23d\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/aca23d","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/aca23d\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/aca23d\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/aca23d\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/aca23d\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/aca23d\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,12,29]],"date-time":"2022-12-29T11:29:40Z","timestamp":1672313380000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/aca23d"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,12,1]]},"references-count":53,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2022,12,29]]},"published-print":{"date-parts":[[2022,12,1]]}},"URL":"https:\/\/doi.org\/10.1088\/2632-2153\/aca23d","relation":{},"ISSN":["2632-2153"],"issn-type":[{"value":"2632-2153","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,12,1]]},"assertion":[{"value":"Self-supervised learning of materials concepts from crystal structures via deep neural networks","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":"2022-07-01","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2022-11-11","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2022-12-29","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}