{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T10:41:36Z","timestamp":1777632096397,"version":"3.51.4"},"reference-count":49,"publisher":"IOP Publishing","issue":"2","license":[{"start":{"date-parts":[[2022,5,5]],"date-time":"2022-05-05T00:00:00Z","timestamp":1651708800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2022,5,5]],"date-time":"2022-05-05T00:00:00Z","timestamp":1651708800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"name":"Oak Ridge National Laboratory - Artificial Intelligence Initiative","award":["3110L111"],"award-info":[{"award-number":["3110L111"]}]},{"DOI":"10.13039\/100000015","name":"U.S. Department of Energy","doi-asserted-by":"crossref","award":["DE-AC05-00OR22725"],"award-info":[{"award-number":["DE-AC05-00OR22725"]}],"id":[{"id":"10.13039\/100000015","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/100006228","name":"Oak Ridge National Laboratory","doi-asserted-by":"crossref","award":["DE-AC05-00OR22725"],"award-info":[{"award-number":["DE-AC05-00OR22725"]}],"id":[{"id":"10.13039\/100006228","id-type":"DOI","asserted-by":"crossref"}]},{"name":"Oak Ridge Leadership Computing Facility","award":["Directorate Discretionary OLCF allocation - CSC457"],"award-info":[{"award-number":["Directorate Discretionary OLCF allocation - CSC457"]}]},{"name":"US Department of Energy"},{"DOI":"10.13039\/100016818","name":"UT-Battelle, LLC","doi-asserted-by":"crossref","id":[{"id":"10.13039\/100016818","id-type":"DOI","asserted-by":"crossref"}]},{"name":"Program of Oak Ridge"}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Mach. Learn.: Sci. Technol."],"published-print":{"date-parts":[[2022,6,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>We introduce a multi-tasking graph convolutional neural network, HydraGNN, to simultaneously predict <jats:italic>both global and atomic<\/jats:italic> physical properties and demonstrate with ferromagnetic materials. We train HydraGNN on an open-source <jats:italic>ab initio<\/jats:italic> density functional theory (DFT) dataset for iron-platinum with a fixed body centered tetragonal lattice structure and fixed volume to simultaneously predict the mixing enthalpy (a global feature of the system), the atomic charge transfer, and the atomic magnetic moment across configurations that span the entire compositional range. By taking advantage of underlying physical correlations between material properties, multi-task learning (MTL) with HydraGNN provides effective training even with modest amounts of data. Moreover, this is achieved with just one architecture instead of three, as required by single-task learning (STL). The first convolutional layers of the HydraGNN architecture are shared by all learning tasks and extract features common to all material properties. The following layers discriminate the features of the different properties, the results of which are fed to the separate heads of the final layer to produce predictions. Numerical results show that HydraGNN effectively captures the relation between the configurational entropy and the material properties over the entire compositional range. Overall, the accuracy of simultaneous MTL predictions is comparable to the accuracy of the STL predictions. In addition, the computational cost of training HydraGNN for MTL is much lower than the original DFT calculations and also lower than training separate STL models for each property.<\/jats:p>","DOI":"10.1088\/2632-2153\/ac6a51","type":"journal-article","created":{"date-parts":[[2022,4,26]],"date-time":"2022-04-26T12:11:12Z","timestamp":1650975072000},"page":"025007","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":16,"title":["Multi-task graph neural networks for simultaneous prediction of global and atomic properties in ferromagnetic systems\n                  <sup>*<\/sup>"],"prefix":"10.1088","volume":"3","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-4980-6924","authenticated-orcid":true,"given":"Massimiliano","family":"Lupo Pasini","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8351-0529","authenticated-orcid":true,"given":"Pei","family":"Zhang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4250-9476","authenticated-orcid":false,"given":"Samuel","family":"Temple Reeve","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6459-6152","authenticated-orcid":false,"given":"Jong","family":"Youl Choi","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"266","published-online":{"date-parts":[[2022,5,5]]},"reference":[{"key":"mlstac6a51bib1","doi-asserted-by":"publisher","first-page":"B864","DOI":"10.1103\/PhysRev.136.B864","article-title":"Inhomogeneous electron gas","volume":"136","author":"Hoenber","year":"1964","journal-title":"Phys. Rev."},{"key":"mlstac6a51bib2","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","author":"Kohn","year":"1965","journal-title":"Phys. Rev."},{"key":"mlstac6a51bib3","author":"Nightingale","year":"1999"},{"key":"mlstac6a51bib4","author":"Hammond","year":"1994"},{"key":"mlstac6a51bib5","doi-asserted-by":"publisher","first-page":"2471","DOI":"10.1103\/PhysRevLett.55.2471","article-title":"Unified approach for molecular dynamics and density-functional theory","volume":"55","author":"Car","year":"1985","journal-title":"Phys. Rev. Lett."},{"key":"mlstac6a51bib6","author":"Marx","year":"2012"},{"key":"mlstac6a51bib7","doi-asserted-by":"publisher","first-page":"2016","DOI":"10.1063\/1.4952607","article-title":"Perspective: web-based machine learning models for real-time screening of thermoelectric materials properties","volume":"4","author":"Gaultois","year":"2016","journal-title":"APL Mater."},{"key":"mlstac6a51bib8","doi-asserted-by":"publisher","first-page":"3405","DOI":"10.1038\/s41467-018-05761-w","article-title":"Accelerated discovery of stable lead-free hybrid organic-inorganic perovskites via machine learning","volume":"9","author":"Lu","year":"2018","journal-title":"Nat. Commun."},{"key":"mlstac6a51bib9","doi-asserted-by":"publisher","first-page":"1120","DOI":"10.1038\/nmat4717","article-title":"Design of efficient molecular organic light-emitting diodes by a high-throughput virtual screening and experimental approach","volume":"15","author":"G\u00f3mez-Bombarelli","year":"2016","journal-title":"Nat. Mater."},{"key":"mlstac6a51bib10","doi-asserted-by":"publisher","DOI":"10.1038\/ncomms11241","article-title":"Accelerated search for materials with targeted properties by adaptive design","volume":"7","author":"Xue","year":"2016","journal-title":"Nat. Commun."},{"key":"mlstac6a51bib11","doi-asserted-by":"publisher","first-page":"59","DOI":"10.1038\/s41524-022-00734-6","article-title":"Recent advances and applications of deep learning methods in materials science","volume":"8","author":"Choudhary","year":"2022","journal-title":"Npj Comput. Mater."},{"key":"mlstac6a51bib12","article-title":"Overcoming data scarcity with transfer learning","author":"Hutchinson","year":"2017"},{"key":"mlstac6a51bib13","doi-asserted-by":"publisher","first-page":"686","DOI":"10.1016\/j.jcp.2018.10.045","article-title":"Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations","volume":"378","author":"Raissi","year":"2019","journal-title":"J. Comput. Phys."},{"key":"mlstac6a51bib14","doi-asserted-by":"publisher","first-page":"2339","DOI":"10.1038\/s41467-019-10343-5","article-title":"Physically informed artificial neural networks for atomistic modeling of materials","volume":"10","author":"Purja","year":"2019","journal-title":"Nat. Commun."},{"key":"mlstac6a51bib15","doi-asserted-by":"publisher","first-page":"422","DOI":"10.1038\/s42254-021-00314-5","article-title":"Physics-informed machine learning","volume":"3","author":"Karniadakis","year":"2021","journal-title":"Nat. Rev. Phys."},{"key":"mlstac6a51bib16","doi-asserted-by":"publisher","first-page":"41","DOI":"10.1016\/b978-1-55860-307-3.50012-5","article-title":"Multitask learning: a knowledge-based source of inductive bias","volume":"48","author":"Caruana","year":"1993","journal-title":"Mach. Learn."},{"key":"mlstac6a51bib17","first-page":"pp 160","article-title":"A unified architecture for natural language processing: deep neural networks with multitask learning","volume":"vol 7","author":"Collobert","year":"2008"},{"key":"mlstac6a51bib18","doi-asserted-by":"publisher","first-page":"2068","DOI":"10.1021\/acs.jcim.7b00146","article-title":"Is multitask deep learning practical for pharma?","volume":"57","author":"Ramsundar","year":"2017","journal-title":"J. Chem. Inf. Model."},{"key":"mlstac6a51bib19","doi-asserted-by":"publisher","DOI":"10.1088\/1361-648x\/abcb10","article-title":"Fast and stable deep-learning predictions of material properties for solid solution alloys","volume":"33","author":"Lupo Pasini","year":"2020","journal-title":"J. Phys.: Condens. Matter."},{"key":"mlstac6a51bib20","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":"mlstac6a51bib21","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":"mlstac6a51bib22","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":"mlstac6a51bib23","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevMaterials.4.063801","article-title":"Developing an improved crystal graph convolutional neural network framework for accelerated materials discovery","volume":"4","author":"Park","year":"2020","journal-title":"Phys. Rev. Mater."},{"key":"mlstac6a51bib24","doi-asserted-by":"publisher","first-page":"411","DOI":"10.1007\/978-3-030-96498-6_24","article-title":"Recurrent multi-task graph convolutional networks for COVID-19 knowledge graph link prediction","volume":"1512","author":"Kim","year":"2022","journal-title":"Driving Scientific and Engineering Discoveries Through the Integration of Experiment, Big Data, and Modeling and Simulation. SMC 2021. Communications in Computer and Information Science"},{"key":"mlstac6a51bib25","article-title":"MT-CGCNN: integrating crystal graph convolutional neural network with multitask learning for material property prediction","author":"Sanyal","year":"2018"},{"key":"mlstac6a51bib26","article-title":"Accurate and scalable multi-element graph neural network force field and molecular dynamics with direct force architecture","author":"Park","year":"2020"},{"key":"mlstac6a51bib27","doi-asserted-by":"publisher","DOI":"10.13139\/OLCF\/1762742","article-title":"FePt binary alloy with 32 atoms - LSMS-3 data","author":"Lupo Pasini","year":"2021"},{"key":"mlstac6a51bib28","article-title":"LSMS","author":"Eisenbach","year":"2017"},{"key":"mlstac6a51bib29","first-page":"pp 8024","volume":"vol 32","author":"Paszke","year":"2019"},{"key":"mlstac6a51bib30","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1912.01703","article-title":"PyTorch: an imperative style, high-performance deep learning ibrary","author":"Paszke","year":"2019"},{"key":"mlstac6a51bib31","article-title":"Fast graph representation learning with PyTorch Geometric","author":"Fey","year":"2019"},{"key":"mlstac6a51bib32","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1903.02428","article-title":"Fast graph representation learning with pyTorch geometric","author":"Fey","year":"2019"},{"key":"mlstac6a51bib33","doi-asserted-by":"publisher","DOI":"10.11578\/dc.20211019.2","article-title":"HydraGNN [Computer Software]","author":"Lupo Pasini","year":"2021"},{"key":"mlstac6a51bib34","author":"Murty","year":"1976"},{"key":"mlstac6a51bib35","doi-asserted-by":"publisher","first-page":"61","DOI":"10.1109\/TNN.2008.2005605","article-title":"The graph neural network model","volume":"20","author":"Scarselli","year":"2009","journal-title":"IEEE Trans. Neural Netw."},{"key":"mlstac6a51bib36","volume":"vol 29","author":"Defferrard","year":"2016"},{"key":"mlstac6a51bib37","article-title":"Principal neighbourhood aggregation for graph nets","author":"Corso"},{"key":"mlstac6a51bib38","first-page":"pp 1025","article-title":"Inductive representation learning on large graphs","author":"Hamilton","year":"2017"},{"key":"mlstac6a51bib39","doi-asserted-by":"publisher","first-page":"79","DOI":"10.1007\/978-3-030-96498-6_5","article-title":"Fast and accurate predictions of total energy for solid solution alloys with graph convolutional neural networks","volume":"1512","author":"Lupo Pasini","year":"2021","journal-title":"Driving Scientific and Engineering Discoveries Through the Integration of Experiment, Big Data, and Modeling and Simulation. SMC 2021. Communications in Computer and Information Science"},{"key":"mlstac6a51bib40","doi-asserted-by":"publisher","first-page":"2","DOI":"10.1016\/j.cpc.2016.07.013","article-title":"GPU acceleration of the locally self-consistent multiple scattering code for first principles calculation of the ground state and statistical physics of materials","volume":"211","author":"Eisenbach","year":"2017","journal-title":"Comput. Phys. Commun."},{"key":"mlstac6a51bib41","doi-asserted-by":"publisher","first-page":"2867","DOI":"10.1103\/PhysRevLett.75.2867","article-title":"Order-N multiple scattering approach to electronic structure calculations","volume":"75","author":"Wang","year":"1995","journal-title":"Phys. Rev. Lett."},{"key":"mlstac6a51bib42","doi-asserted-by":"publisher","first-page":"218","DOI":"10.1016\/j.commatsci.2012.02.005","article-title":"AFLOW: an automatic framework for high-throughput materials discovery","volume":"58","author":"Curtarolo","year":"2012","journal-title":"Comput. Mater. Sci."},{"key":"mlstac6a51bib43","doi-asserted-by":"publisher","first-page":"0","DOI":"10.1063\/1.4812323","article-title":"Commentary: the materials project: a materials genome approach to accelerating materials innovation","volume":"1","author":"Jain","year":"2013","journal-title":"APL Mater."},{"key":"mlstac6a51bib44","doi-asserted-by":"publisher","first-page":"1501","DOI":"10.1007\/s11837-013-0755-4","article-title":"Materials design and discovery with high-throughput density functional theory: the Open Quantum Materials Database (OQMD)","volume":"65","author":"Saal","year":"2013","journal-title":"JOM"},{"key":"mlstac6a51bib45","article-title":"Adam: a method for stochastic optimization","author":"Kingma","year":"2017"},{"key":"mlstac6a51bib46","doi-asserted-by":"publisher","first-page":"2355","DOI":"10.1088\/0022-3727\/31\/19\/003","article-title":"Magnetic alloys, their electronic structure and micromagnetic and microstructural models","volume":"31","author":"Staunton","year":"1998","journal-title":"J. Phys. D: Appl. Phys"},{"key":"mlstac6a51bib47","doi-asserted-by":"publisher","first-page":"7884","DOI":"10.1038\/s41598-018-25978-5","article-title":"Magnetism and high magnetic-field-induced stability of alloy carbides in fe-based materials","volume":"8","author":"Hou","year":"2018","journal-title":"Sci. Rep."},{"key":"mlstac6a51bib48","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-019-44351-8","article-title":"Combinatorial evaluation of phase formation and magnetic properties of femncocral high entropy alloy thin film library","volume":"9","author":"Marshal","year":"2019","journal-title":"Sci. Rep."},{"key":"mlstac6a51bib49","doi-asserted-by":"publisher","DOI":"10.1016\/j.parco.2021.102788","article-title":"A scalable constructive algorithm for the optimization of neural network architectures","volume":"104\u2013105","author":"Lupo Pasini","year":"2021","journal-title":"Parallel Comput."}],"container-title":["Machine Learning: Science and Technology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac6a51","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac6a51\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac6a51","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac6a51\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac6a51\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac6a51\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac6a51\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac6a51\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,5,5]],"date-time":"2022-05-05T07:41:38Z","timestamp":1651736498000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac6a51"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,5,5]]},"references-count":49,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2022,5,5]]},"published-print":{"date-parts":[[2022,6,1]]}},"URL":"https:\/\/doi.org\/10.1088\/2632-2153\/ac6a51","relation":{},"ISSN":["2632-2153"],"issn-type":[{"value":"2632-2153","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,5,5]]},"assertion":[{"value":"Multi-task graph neural networks for simultaneous prediction of global and atomic properties in ferromagnetic systems\n                  *","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":"2022-02-07","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2022-04-25","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2022-05-05","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}