{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,2]],"date-time":"2026-05-02T03:48:26Z","timestamp":1777693706154,"version":"3.51.4"},"reference-count":59,"publisher":"IOP Publishing","issue":"3","license":[{"start":{"date-parts":[[2023,9,12]],"date-time":"2023-09-12T00:00:00Z","timestamp":1694476800000},"content-version":"vor","delay-in-days":11,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2023,9,12]],"date-time":"2023-09-12T00:00:00Z","timestamp":1694476800000},"content-version":"tdm","delay-in-days":11,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"DOI":"10.13039\/100021130","name":"Bundesministerium f\u00fcr Wirtschaft und Klimaschutz","doi-asserted-by":"crossref","award":["KK5139001AP0"],"award-info":[{"award-number":["KK5139001AP0"]}],"id":[{"id":"10.13039\/100021130","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100003542","name":"Ministerium f\u00fcr Wissenschaft, Forschung und Kunst Baden-W\u00fcrttemberg","doi-asserted-by":"crossref","award":["bwHPC"],"award-info":[{"award-number":["bwHPC"]}],"id":[{"id":"10.13039\/501100003542","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100002347","name":"Bundesministerium f\u00fcr Bildung und Forschung","doi-asserted-by":"crossref","award":["HoreKa project AI-MAT"],"award-info":[{"award-number":["HoreKa project AI-MAT"]}],"id":[{"id":"10.13039\/501100002347","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":[[2023,9,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Accurate prediction of the ionization potential and electron affinity energies of small molecules are important for many applications. Density functional theory (DFT) is computationally inexpensive, but can be very inaccurate for frontier orbital energies or ionization energies. The GW method is sufficiently accurate for many relevant applications, but much more expensive than DFT. Here we study how we can learn to predict orbital energies with GW accuracy using machine learning (ML) on molecular graphs and fingerprints using an interpretable delta-learning approach. ML models presented here can be used to predict quasiparticle energies of small organic molecules even beyond the size of the molecules used for training. We furthermore analyze the learned DFT-to-GW corrections by mapping them to specific localized fragments of the molecules, in order to develop an intuitive interpretation of the learned corrections, and thus to better understand DFT errors.<\/jats:p>","DOI":"10.1088\/2632-2153\/acf545","type":"journal-article","created":{"date-parts":[[2023,8,30]],"date-time":"2023-08-30T22:40:49Z","timestamp":1693435249000},"page":"035045","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":5,"title":["Interpretable delta-learning of GW quasiparticle energies from GGA-DFT"],"prefix":"10.1088","volume":"4","author":[{"given":"Artem","family":"Fediai","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7052-696X","authenticated-orcid":true,"given":"Patrick","family":"Reiser","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jorge Enrique Olivares","family":"Pe\u00f1a","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Wolfgang","family":"Wenzel","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4465-1465","authenticated-orcid":true,"given":"Pascal","family":"Friederich","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"266","published-online":{"date-parts":[[2023,9,12]]},"reference":[{"key":"mlstacf545bib1","doi-asserted-by":"publisher","first-page":"191","DOI":"10.1038\/nmat3568","article-title":"The high-throughput highway to computational materials design","volume":"12","author":"Curtarolo","year":"2013","journal-title":"Nat. Mater."},{"key":"mlstacf545bib2","doi-asserted-by":"publisher","DOI":"10.1002\/adma.201703063","article-title":"Controlling molecular doping in organic semiconductors","volume":"29","author":"Jacobs","year":"2017","journal-title":"Adv. Mater."},{"key":"mlstacf545bib3","doi-asserted-by":"publisher","first-page":"3750","DOI":"10.1021\/acs.jctc.1c00191","article-title":"Analyzing dynamical disorder for charge transport in organic semiconductors via machine learning","volume":"17","author":"Reiser","year":"2021","journal-title":"J. Chem. Theory Comput."},{"key":"mlstacf545bib4","doi-asserted-by":"publisher","first-page":"3727","DOI":"10.1021\/acs.jctc.1c00036","article-title":"Computing charging and polarization energies of small organic molecules embedded into amorphous materials with quantum accuracy","volume":"17","author":"Armleder","year":"2021","journal-title":"J. Chem. Theory Comput."},{"key":"mlstacf545bib5","doi-asserted-by":"publisher","first-page":"630","DOI":"10.1002\/sdtp.13946","article-title":"43\u20133: ab-initio simulation of doped injection layers","volume":"vol 51","author":"Symalla","year":"2020","journal-title":"SID Symp. Digest of Technical Papers"},{"key":"mlstacf545bib6","doi-asserted-by":"publisher","first-page":"56","DOI":"10.1016\/j.commatsci.2015.02.050","article-title":"The electrolyte genome project: a big data approach in battery materials discovery","volume":"103","author":"Qu","year":"2015","journal-title":"Comput. Mater. Sci."},{"key":"mlstacf545bib7","doi-asserted-by":"publisher","first-page":"16495","DOI":"10.1039\/C8TA05922E","article-title":"Influence of dopant size and electron affinity on the electrical conductivity and thermoelectric properties of a series of conjugated polymers","volume":"6","author":"Liang","year":"2018","journal-title":"J. Mater. Chem. A"},{"key":"mlstacf545bib8","doi-asserted-by":"publisher","first-page":"8481","DOI":"10.1021\/acscatal.9b01775","article-title":"Beyond density functional theory: the multiconfigurational approach to model heterogeneous catalysis","volume":"9","author":"Gaggioli","year":"2019","journal-title":"ACS Catal."},{"key":"mlstacf545bib9","doi-asserted-by":"publisher","DOI":"10.1016\/j.molstruc.2020.129776","article-title":"Screening donor and acceptor groups for organic azo-based dyes for dye sensitized solar cells","volume":"1228","author":"Biswal","year":"2021","journal-title":"J. Mol. Struct."},{"key":"mlstacf545bib10","doi-asserted-by":"publisher","first-page":"3324","DOI":"10.1039\/D1TC00002K","article-title":"Novel thermally activated delayed fluorescence materials by high-throughput virtual screening: going beyond donor\u2013acceptor design","volume":"9","author":"Zhao","year":"2021","journal-title":"J. Mater. Chem. C"},{"key":"mlstacf545bib11","doi-asserted-by":"publisher","DOI":"10.1063\/1.4704546","article-title":"Perspective on density functional theory","volume":"136","author":"Burke","year":"2012","journal-title":"J. Chem. Phys."},{"key":"mlstacf545bib12","doi-asserted-by":"publisher","first-page":"1554","DOI":"10.1021\/jp061633o","article-title":"Comparison of DFT methods for molecular orbital eigenvalue calculations","volume":"111","author":"Zhang","year":"2007","journal-title":"J. Phys. Chem. A"},{"key":"mlstacf545bib13","doi-asserted-by":"publisher","first-page":"291","DOI":"10.1103\/RevModPhys.79.291","article-title":"Coupled-cluster theory in quantum chemistry","volume":"79","author":"Bartlett","year":"2007","journal-title":"Rev. Mod. Phys."},{"key":"mlstacf545bib14","doi-asserted-by":"publisher","first-page":"108","DOI":"10.1021\/cr200137a","article-title":"Multiconfiguration self-consistent field and multireference configuration interaction methods and applications","volume":"112","author":"Szalay","year":"2012","journal-title":"Chem. Rev."},{"key":"mlstacf545bib15","doi-asserted-by":"publisher","first-page":"2421","DOI":"10.1103\/PhysRevA.49.2421","article-title":"Exchange-correlation potential with correct asymptotic behavior","volume":"49","author":"van Leeuwen","year":"1994","journal-title":"Phys. Rev. A"},{"key":"mlstacf545bib16","doi-asserted-by":"publisher","first-page":"237","DOI":"10.1088\/0034-4885\/61\/3\/002","article-title":"The GW method","volume":"61","author":"Aryasetiawan","year":"1998","journal-title":"Rep. Prog. Phys."},{"key":"mlstacf545bib17","doi-asserted-by":"publisher","DOI":"10.5445\/IR\/1000052904","article-title":"Quasiparticle self-consistent GW-approximation for molecules. Calculation of single-particle excitation energies for molecules","author":"Kaplan","year":"2015"},{"key":"mlstacf545bib18","doi-asserted-by":"publisher","first-page":"232","DOI":"10.1021\/ct300648t","article-title":"The GW-method for quantum chemistry applications: theory and implementation","volume":"9","author":"van Setten","year":"2013","journal-title":"J. Chem. Theory Comput."},{"key":"mlstacf545bib19","doi-asserted-by":"publisher","first-page":"3623","DOI":"10.1021\/acs.jctc.6b00380","article-title":"GW in the Gaussian and plane waves scheme with application to linear acenes","volume":"12","author":"Wilhelm","year":"2016","journal-title":"J. Chem. Theory Comput."},{"key":"mlstacf545bib20","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevB.87.235132","article-title":"Quasiparticle GW calculations for solids, molecules, and two-dimensional materials","volume":"87","author":"H\u00fcser","year":"2013","journal-title":"Phys. Rev. B"},{"key":"mlstacf545bib21","doi-asserted-by":"publisher","first-page":"A796","DOI":"10.1103\/PhysRev.139.A796","article-title":"New method for calculating the one-particle Green\u2019s function with application to the electron-gas problem","volume":"139","author":"Hedin","year":"1965","journal-title":"Phys. Rev."},{"key":"mlstacf545bib22","doi-asserted-by":"publisher","first-page":"225","DOI":"10.1002\/9781119356059.ch5","article-title":"Machine learning, quantum chemistry, and chemical space","author":"Ramakrishnan","year":"2017"},{"key":"mlstacf545bib23","doi-asserted-by":"publisher","first-page":"1032","DOI":"10.1002\/qua.24890","article-title":"Constructing high-dimensional neural network potentials: a tutorial review","volume":"115","author":"Behler","year":"2015","journal-title":"Int. J. Quantum Chem."},{"key":"mlstacf545bib24","doi-asserted-by":"publisher","first-page":"1058","DOI":"10.1002\/qua.24954","article-title":"Machine learning for quantum mechanics in a nutshell","volume":"115","author":"Rupp","year":"2015","journal-title":"Int. J. Quantum Chem."},{"key":"mlstacf545bib25","doi-asserted-by":"publisher","first-page":"93","DOI":"10.1038\/s43246-022-00315-6","article-title":"Graph neural networks for materials science and chemistry","volume":"3","author":"Reiser","year":"2022","journal-title":"Commun. Mater."},{"key":"mlstacf545bib26","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":"mlstacf545bib27","article-title":"How powerful are graph neural networks?","author":"Xu","year":"2019"},{"key":"mlstacf545bib28","article-title":"Neural message passing for quantum chemistry","volume":"vol 70","author":"Gilmer","year":"2017"},{"key":"mlstacf545bib29","article-title":"Semi-supervised classification with graph convolutional networks","author":"Kipf","year":"2017"},{"key":"mlstacf545bib30","doi-asserted-by":"publisher","first-page":"2087","DOI":"10.1021\/acs.jctc.5b00099","article-title":"Big data meets quantum chemistry approximations: the \u0394-machine learning approach","volume":"11","author":"Ramakrishnan","year":"2015","journal-title":"J. Chem. Theory Comput."},{"key":"mlstacf545bib31","doi-asserted-by":"publisher","first-page":"5804","DOI":"10.1021\/acs.jpca.0c01777","article-title":"Quantum-chemically informed machine learning: prediction of energies of organic molecules with 10\u201314 non-hydrogen atoms","volume":"124","author":"Dandu","year":"2020","journal-title":"J. Phys. Chem. A"},{"key":"mlstacf545bib32","doi-asserted-by":"publisher","first-page":"4891","DOI":"10.1021\/acs.jctc.1c00520","article-title":"Machine learning of quasiparticle energies in molecules and clusters","volume":"17","author":"\u00c7aylak","year":"2021","journal-title":"J. Chem. Theory Comput."},{"key":"mlstacf545bib33","doi-asserted-by":"publisher","first-page":"2864","DOI":"10.1021\/ci300415d","article-title":"Enumeration of 166 billion organic small molecules in the Chemical Universe Database GDB-17","volume":"52","author":"Ruddigkeit","year":"2012","journal-title":"J. Chem. Inf. Model."},{"key":"mlstacf545bib34","doi-asserted-by":"publisher","DOI":"10.1038\/sdata.2014.22","article-title":"Quantum chemistry structures and properties of 134 kilo molecules","volume":"1","author":"Ramakrishnan","year":"2014","journal-title":"Sci. Data"},{"key":"mlstacf545bib35","doi-asserted-by":"crossref","DOI":"10.1038\/s41597-023-02486-4","article-title":"Accurate GW frontier orbital energies of 134 kilo molecules","author":"Fediai","year":"2023"},{"key":"mlstacf545bib36","article-title":"Fast and uncertainty-aware directional message passing for non-equilibrium molecules","author":"Klicpera","year":"2020"},{"key":"mlstacf545bib37","doi-asserted-by":"publisher","first-page":"52138","DOI":"10.1109\/ACCESS.2018.2870052","article-title":"Peeking inside the black-box: a survey on explainable artificial intelligence (XAI)","volume":"6","author":"Adadi","year":"2018","journal-title":"IEEE Access"},{"key":"mlstacf545bib38","doi-asserted-by":"publisher","first-page":"82","DOI":"10.1016\/j.inffus.2019.12.012","article-title":"Explainable artificial intelligence (XAI): concepts, taxonomies, opportunities and challenges toward responsible AI","volume":"58","author":"Barredo Arrieta","year":"2020","journal-title":"Inf. Fusion"},{"key":"mlstacf545bib39","article-title":"RDKit: open-source cheminformatics software","author":"Landrum","year":"2016"},{"key":"mlstacf545bib40","doi-asserted-by":"publisher","DOI":"10.1063\/5.0007045","article-title":"CP2K: an electronic structure and molecular dynamics software package\u2014quickstep: efficient and accurate electronic structure calculations","volume":"152","author":"K\u00fchne","year":"2020","journal-title":"J. Chem. Phys."},{"key":"mlstacf545bib41","doi-asserted-by":"publisher","first-page":"267","DOI":"10.1007\/s00214-005-0635-2","article-title":"Estimating the Hartree\u2014Fock limit from finite basis set calculations","volume":"113","author":"Jensen","year":"2005","journal-title":"Theor. Chem. Acc."},{"key":"mlstacf545bib42","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":"mlstacf545bib43","doi-asserted-by":"publisher","first-page":"945","DOI":"10.1038\/s41557-020-0527-z","article-title":"Quantum machine learning using atom-in-molecule-based fragments selected on the fly","volume":"12","author":"Huang","year":"2020","journal-title":"Nat. Chem."},{"key":"mlstacf545bib44","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":"mlstacf545bib45","first-page":"2825","article-title":"Scikit-learn: machine learning in Python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"J. Mach. Learn. Res."},{"key":"mlstacf545bib46","doi-asserted-by":"publisher","DOI":"10.5281\/zenodo.817332","article-title":"QML: a Python toolkit for quantum machine learning","author":"Christensen","year":"2017"},{"key":"mlstacf545bib47","doi-asserted-by":"publisher","DOI":"10.1063\/1.5086105","article-title":"Chemical diversity in molecular orbital energy predictions with kernel ridge regression","volume":"150","author":"Stuke","year":"2019","journal-title":"J. Chem. Phys."},{"key":"mlstacf545bib48","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":"mlstacf545bib49","article-title":"Directional message passing for molecular graphs","author":"Klicpera","year":"2020"},{"key":"mlstacf545bib50","doi-asserted-by":"publisher","DOI":"10.1016\/j.simpa.2021.100095","article-title":"Graph neural networks in tensorflow-keras with raggedtensor representation (Kgcnn)","volume":"9","author":"Reiser","year":"2021","journal-title":"Softw. Impacts"},{"key":"mlstacf545bib51","doi-asserted-by":"publisher","DOI":"10.1088\/1361-648X\/29\/10\/103003","article-title":"Accurate description of the electronic structure of organic semiconductors by GW methods","volume":"29","author":"Marom","year":"2017","journal-title":"J. Phys.: Condens. Matter"},{"key":"mlstacf545bib52","doi-asserted-by":"publisher","first-page":"2696","DOI":"10.1021\/acs.jcim.2c00485","article-title":"Delta machine learning to improve scoring-ranking-screening performances of protein\u2013ligand scoring functions","volume":"62","author":"Yang","year":"2022","journal-title":"J. Chem. Inf. Model."},{"key":"mlstacf545bib53","doi-asserted-by":"publisher","first-page":"342","DOI":"10.1021\/ci600423u","article-title":"Virtual exploration of the chemical universe up to 11 atoms of C, N, O, F: assembly of 26.4 million structures (110.9 million stereoisomers) and analysis for new ring systems, stereochemistry, physicochemical properties, compound classes, and drug discovery","volume":"47","author":"Fink","year":"2007","journal-title":"J. Chem. Inf. Model."},{"key":"mlstacf545bib54","doi-asserted-by":"publisher","first-page":"8732","DOI":"10.1021\/ja902302h","article-title":"970 million druglike small molecules for virtual screening in the chemical universe database GDB-13","volume":"131","author":"Blum","year":"2009","journal-title":"J. Am. Chem. Soc."},{"key":"mlstacf545bib55","article-title":"Gaussian 09, revision D.01 (Gaussian, Inc.)","author":"Frisch","year":"2009"},{"key":"mlstacf545bib56","doi-asserted-by":"publisher","DOI":"10.1063\/5.0004635","article-title":"TURBOMOLE: modular program suite for ab initio quantum-chemical and condensed-matter simulations","volume":"152","author":"Balasubramani","year":"2020","journal-title":"J. Chem. Phys."},{"key":"mlstacf545bib57","doi-asserted-by":"publisher","first-page":"165","DOI":"10.1016\/0009-2614(89)85118-8","article-title":"Electronic structure calculations on workstation computers: the program system turbomole","volume":"162","author":"Ahlrichs","year":"1989","journal-title":"Chem. Phys. Lett."},{"key":"mlstacf545bib58","doi-asserted-by":"publisher","first-page":"10037","DOI":"10.1021\/acs.chemrev.0c00868","article-title":"Four generations of high-dimensional neural network potentials","volume":"121","author":"Behler","year":"2021","journal-title":"Chem. Rev."},{"key":"mlstacf545bib59","doi-asserted-by":"publisher","DOI":"10.1063\/1.5023802","article-title":"less is more: sampling chemical space with active learning","volume":"148","author":"Smith","year":"2018","journal-title":"J. Chem. Phys."}],"container-title":["Machine Learning: Science and Technology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acf545","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acf545\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acf545","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acf545\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acf545\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acf545\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acf545\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acf545\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,9,15]],"date-time":"2023-09-15T14:00:52Z","timestamp":1694786452000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acf545"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,9,1]]},"references-count":59,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2023,9,12]]},"published-print":{"date-parts":[[2023,9,1]]}},"URL":"https:\/\/doi.org\/10.1088\/2632-2153\/acf545","relation":{},"ISSN":["2632-2153"],"issn-type":[{"value":"2632-2153","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,9,1]]},"assertion":[{"value":"Interpretable delta-learning of GW quasiparticle energies from GGA-DFT","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 2023 The Author(s). Published by IOP Publishing Ltd","name":"copyright_information","label":"Copyright Information"},{"value":"2023-03-30","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2023-08-30","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2023-09-12","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}