{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,23]],"date-time":"2026-04-23T07:01:02Z","timestamp":1776927662358,"version":"3.51.2"},"reference-count":48,"publisher":"IOP Publishing","issue":"3","license":[{"start":{"date-parts":[[2022,9,20]],"date-time":"2022-09-20T00:00:00Z","timestamp":1663632000000},"content-version":"vor","delay-in-days":19,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2022,9,20]],"date-time":"2022-09-20T00:00:00Z","timestamp":1663632000000},"content-version":"tdm","delay-in-days":19,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"DOI":"10.13039\/100006134","name":"Office of Energy Efficiency and Renewable Energy","doi-asserted-by":"crossref","award":["DE-0008822"],"award-info":[{"award-number":["DE-0008822"]}],"id":[{"id":"10.13039\/100006134","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/100006151","name":"Basic Energy Sciences","doi-asserted-by":"crossref","award":["DE-FOA-0001912"],"award-info":[{"award-number":["DE-FOA-0001912"]}],"id":[{"id":"10.13039\/100006151","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":[[2022,9,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Progress towards the energy breakthroughs needed to combat climate change can be significantly accelerated through the efficient simulation of atomistic systems. However, simulation techniques based on first principles, such as density functional theory (DFT), are limited in their practical use due to their high computational expense. Machine learning approaches have the potential to approximate DFT in a computationally efficient manner, which could dramatically increase the impact of computational simulations on real-world problems. However, they are limited by their accuracy and the cost of generating labeled data. Here, we present an online active learning framework for accelerating the simulation of atomic systems efficiently and accurately by incorporating prior physical information learned by large-scale pre-trained graph neural network models from the Open Catalyst Project. Accelerating these simulations enables useful data to be generated more cheaply, allowing better models to be trained and more atomistic systems to be screened. We also present a method of comparing local optimization techniques on the basis of both their speed and accuracy. Experiments on 30 benchmark adsorbate-catalyst systems show that our method of transfer learning to incorporate prior information from pre-trained models accelerates simulations by reducing the number of DFT calculations by 91%, while meeting an accuracy threshold of 0.02\u2009eV 93% of the time. Finally, we demonstrate a technique for leveraging the interactive functionality built in to Vienna <jats:italic>ab initio<\/jats:italic> Simulation Package (VASP) to efficiently compute single point calculations within our online active learning framework without the significant startup costs. This allows VASP to work in tandem with our framework while requiring 75% fewer self-consistent cycles than conventional single point calculations. The online active learning implementation, and examples using the VASP interactive code, are available in the open source <jats:italic>FINETUNA<\/jats:italic> package on Github.<\/jats:p>","DOI":"10.1088\/2632-2153\/ac8fe0","type":"journal-article","created":{"date-parts":[[2022,9,6]],"date-time":"2022-09-06T22:45:13Z","timestamp":1662504313000},"page":"03LT01","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":30,"title":["FINETUNA: fine-tuning accelerated molecular simulations"],"prefix":"10.1088","volume":"3","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-5296-9177","authenticated-orcid":false,"given":"Joseph","family":"Musielewicz","sequence":"first","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8587-8610","authenticated-orcid":false,"given":"Xiaoxiao","family":"Wang","sequence":"additional","affiliation":[]},{"given":"Tian","family":"Tian","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9401-4918","authenticated-orcid":true,"given":"Zachary","family":"Ulissi","sequence":"additional","affiliation":[]}],"member":"266","published-online":{"date-parts":[[2022,9,20]]},"reference":[{"key":"mlstac8fe0bib1","doi-asserted-by":"publisher","first-page":"517","DOI":"10.1021\/acs.accounts.6b00510","article-title":"Heterogeneous catalysis: a central science for a sustainable future","volume":"50","author":"Friend","year":"2017","journal-title":"Acc. Chem. Res."},{"key":"mlstac8fe0bib2","doi-asserted-by":"publisher","first-page":"5245","DOI":"10.1021\/acscatal.9b00994","article-title":"Homogeneous, heterogeneous and biological catalysts for electrochemical N2 reduction toward NH3 under ambient conditions","volume":"9","author":"Liu","year":"2019","journal-title":"ACS Catal."},{"key":"mlstac8fe0bib3","doi-asserted-by":"publisher","first-page":"6635","DOI":"10.1021\/acssuschemeng.8b00423","article-title":"Heterogeneous catalytic reactor for hydrogen production from formic acid and its use in polymer electrolyte fuel cells","volume":"6","author":"Yuranov","year":"2018","journal-title":"ACS Sustain. Chem. Eng."},{"key":"mlstac8fe0bib4","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41467-019-13638-9","article-title":"CO2 hydrogenation to high-value products via heterogeneous catalysis","volume":"10","author":"Ye","year":"2019","journal-title":"Nat. Commun."},{"key":"mlstac8fe0bib5","doi-asserted-by":"publisher","first-page":"14147","DOI":"10.1021\/acscatal.0c04273","article-title":"Advances in the design of heterogeneous catalysts and thermocatalytic processes for CO2 utilization","volume":"10","author":"De","year":"2020","journal-title":"ACS Catal."},{"key":"mlstac8fe0bib6","doi-asserted-by":"publisher","first-page":"490","DOI":"10.1038\/s41929-018-0092-7","article-title":"Catalysts for nitrogen reduction to ammonia","volume":"1","author":"Foster","year":"2018","journal-title":"Nat. Catal."},{"key":"mlstac8fe0bib7","doi-asserted-by":"publisher","first-page":"37","DOI":"10.1021\/jacs.7b08891","article-title":"Titanium-based hydrides as heterogeneous catalysts for ammonia synthesis","volume":"139","author":"Kobayashi","year":"2017","journal-title":"J. Am. Chem. Soc."},{"key":"mlstac8fe0bib8","doi-asserted-by":"publisher","first-page":"5838","DOI":"10.1002\/cctc.202001141","article-title":"Recent advances in heterogeneous catalysis for ammonia synthesis","volume":"12","author":"Marakatti","year":"2020","journal-title":"ChemCatChem"},{"key":"mlstac8fe0bib9","doi-asserted-by":"publisher","first-page":"11174","DOI":"10.1039\/D0CP00972E","article-title":"High-throughput experimentation meets artificial intelligence: a new pathway to catalyst discovery","volume":"22","author":"McCullough","year":"2020","journal-title":"Phys. Chem. Chem. Phys."},{"key":"mlstac8fe0bib10","doi-asserted-by":"publisher","first-page":"206","DOI":"10.1016\/j.jcat.2004.02.034","article-title":"The Br\u00f8nsted\u2013Evans\u2013Polanyi relation and the volcano curve in heterogeneous catalysis","volume":"224","author":"Bligaard","year":"2004","journal-title":"J. Catal."},{"key":"mlstac8fe0bib11","doi-asserted-by":"publisher","first-page":"12974","DOI":"10.1021\/jp960669l","article-title":"Density functional theory of electronic structure","volume":"100","author":"Kohn","year":"1996","journal-title":"J. Phys. Chem."},{"key":"mlstac8fe0bib12","doi-asserted-by":"publisher","first-page":"2311","DOI":"10.1002\/aic.16198","article-title":"Machine learning for heterogeneous catalyst design and discovery","volume":"64","author":"Goldsmith","year":"2018","journal-title":"AIChE J."},{"key":"mlstac8fe0bib13","doi-asserted-by":"publisher","first-page":"157","DOI":"10.1021\/acs.chemmater.9b03043","article-title":"Enabling catalyst discovery through machine learning and high-throughput experimentation","volume":"32","author":"Williams","year":"2020","journal-title":"Chem. Mater."},{"key":"mlstac8fe0bib14","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevB.100.104103","article-title":"Local Bayesian optimizer for atomic structures","volume":"100","author":"Del R\u00edo","year":"2019","journal-title":"Phys. Rev. B"},{"key":"mlstac8fe0bib15","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.98.146401","article-title":"Generalized neural-network representation of high-dimensional potential-energy surfaces","volume":"98","author":"Behler","year":"2007","journal-title":"Phys. Rev. Lett."},{"key":"mlstac8fe0bib16","doi-asserted-by":"publisher","first-page":"3192","DOI":"10.1039\/C6SC05720A","article-title":"ANI-1: an extensible neural network potential with DFT accuracy at force field computational cost","volume":"8","author":"Smith","year":"2017","journal-title":"Chem. Sci."},{"key":"mlstac8fe0bib17","doi-asserted-by":"publisher","first-page":"3408","DOI":"10.1021\/acs.jcim.0c00451","article-title":"TorchANI: a free and open source PyTorch-based deep learning implementation of the ANI neural network potentials","volume":"60","author":"Gao","year":"2020","journal-title":"J. Chem. Inf. Model."},{"key":"mlstac8fe0bib18","doi-asserted-by":"publisher","first-page":"4192","DOI":"10.1021\/acs.jctc.0c00121","article-title":"Extending the applicability of the ANI deep learning molecular potential to sulfur and halogens","volume":"16","author":"Devereux","year":"2020","journal-title":"J. Chem. Theory Comput."},{"key":"mlstac8fe0bib19","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41597-020-0473-z","article-title":"The ANI-1ccx and ANI-1x data sets, coupled-cluster and density functional theory properties for molecules","volume":"7","author":"Smith","year":"2020","journal-title":"Sci. Data"},{"key":"mlstac8fe0bib20","article-title":"Directional message passing for molecular graphs","author":"Klicpera","year":"2020"},{"key":"mlstac8fe0bib21","article-title":"Fast and uncertainty-aware directional message passing for non-equilibrium molecules","author":"Klicpera","year":"2020"},{"key":"mlstac8fe0bib22","article-title":"Rotation invariant graph neural networks using spin convolutions","author":"Shuaibi","year":"2021"},{"key":"mlstac8fe0bib23","article-title":"GemNet: Universal directional graph neural networks for molecules","author":"Gasteiger","year":"2021"},{"key":"mlstac8fe0bib24","doi-asserted-by":"publisher","first-page":"6059","DOI":"10.1021\/acscatal.0c04525","article-title":"Open catalyst 2020 (OC20) dataset and community challenges","volume":"11","author":"Chanussot","year":"2021","journal-title":"ACS Catal."},{"key":"mlstac8fe0bib25","doi-asserted-by":"publisher","first-page":"8920","DOI":"10.1039\/D0CC03512B","article-title":"Active learning and neural network potentials accelerate molecular screening of ether-based solvate ionic liquids","volume":"56","author":"Wang","year":"2020","journal-title":"Chem. Commun."},{"key":"mlstac8fe0bib26","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41524-019-0153-8","article-title":"Active learning in materials science with emphasis on adaptive sampling using uncertainties for targeted design","volume":"5","author":"Lookman","year":"2019","journal-title":"npj Comput. Mater."},{"key":"mlstac8fe0bib27","doi-asserted-by":"publisher","first-page":"696","DOI":"10.1038\/s41929-018-0142-1","article-title":"Active learning across intermetallics to guide discovery of electrocatalysts for CO2 reduction and H2 evolution","volume":"1","author":"Tran","year":"2018","journal-title":"Nat. Catal."},{"key":"mlstac8fe0bib28","doi-asserted-by":"publisher","first-page":"178","DOI":"10.1038\/s41586-020-2242-8","article-title":"Accelerated discovery of CO2 electrocatalysts using active machine learning","volume":"581","author":"Zhong","year":"2020","journal-title":"Nature"},{"key":"mlstac8fe0bib29","doi-asserted-by":"crossref","DOI":"10.21203\/rs.3.rs-1178160\/v1","article-title":"Active learning of reactive Bayesian force fields: application to heterogeneous hydrogen-platinum catalysis dynamics","author":"Vandermause","year":"2021"},{"key":"mlstac8fe0bib30","doi-asserted-by":"publisher","first-page":"20","DOI":"10.1038\/s41524-020-0283-z","article-title":"On-the-fly active learning of interpretable Bayesian force fields for atomistic rare events","volume":"6","author":"Vandermause","year":"2020","journal-title":"npj Comput. Mater."},{"key":"mlstac8fe0bib31","doi-asserted-by":"publisher","DOI":"10.1063\/5.0049665","article-title":"Machine-learning accelerated geometry optimization in molecular simulation","volume":"154","author":"Yang","year":"2021","journal-title":"J. Chem. Phys."},{"key":"mlstac8fe0bib32","doi-asserted-by":"publisher","DOI":"10.1088\/2632-2153\/abcc44","article-title":"Enabling robust offline active learning for machine learning potentials using simple physics-based priors","volume":"2","author":"Shuaibi","year":"2021","journal-title":"Mach. Learn.: Sci. Technol."},{"key":"mlstac8fe0bib33","doi-asserted-by":"crossref","DOI":"10.1088\/2632-2153\/ac8fe0","article-title":"FINETUNA: fine-tuning accelerated molecular simulations","author":"Musielewicz","year":"2022"},{"key":"mlstac8fe0bib34","doi-asserted-by":"publisher","first-page":"15","DOI":"10.1016\/0927-0256(96)00008-0","article-title":"Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set","volume":"6","author":"Kresse","year":"1996","journal-title":"Comput. Mater. Sci."},{"key":"mlstac8fe0bib35","doi-asserted-by":"publisher","first-page":"222","DOI":"10.1016\/0022-3093(95)00355-X","article-title":"Ab initio molecular dynamics for liquid metals","volume":"192\u2013193","author":"Kresse","year":"1995","journal-title":"J. Non-Cryst. Solids"},{"key":"mlstac8fe0bib36","doi-asserted-by":"publisher","first-page":"14251","DOI":"10.1103\/PhysRevB.49.14251","article-title":"Ab initio molecular-dynamics simulation of the liquid-metalamorphous-semiconductor transition in germanium","volume":"49","author":"Kresse","year":"1994","journal-title":"Phys. Rev. B"},{"key":"mlstac8fe0bib37","doi-asserted-by":"publisher","first-page":"11169","DOI":"10.1103\/PhysRevB.54.11169","article-title":"Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set","volume":"54","author":"Kresse","year":"1996","journal-title":"Phys. Rev. B"},{"key":"mlstac8fe0bib38","doi-asserted-by":"publisher","DOI":"10.1088\/1361-648X\/aa680e","article-title":"The atomic simulation environment\u2014a Python library for working with atoms","volume":"29","author":"Hjorth Larsen","year":"2017","journal-title":"J. Phys.: Condens. Matter"},{"key":"mlstac8fe0bib39","article-title":"VASPInteractive: interactive VASP calculator","author":"Tian","year":"2022"},{"key":"mlstac8fe0bib40","first-page":"pp 194","article-title":"How to fine-tune BERT for text classification?","author":"Sun","year":"2019"},{"key":"mlstac8fe0bib41","article-title":"Active learning literature survey","volume":"vol 1648","author":"Settles","year":"2009"},{"key":"mlstac8fe0bib42","doi-asserted-by":"crossref","DOI":"10.1088\/2632-2153\/ac8fe0","article-title":"FINETUNA: fine-tuning accelerated molecular simulations manuscript","author":"Musielewicz","year":"2022"},{"key":"mlstac8fe0bib43","doi-asserted-by":"publisher","first-page":"11169","DOI":"10.1103\/PhysRevB.54.11169","article-title":"Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set","volume":"54","author":"Kresse","year":"1996","journal-title":"Phys. Rev. B"},{"key":"mlstac8fe0bib44","article-title":"NELMDL: non-self consistent steps in electronic minimization from The VASP Manual","author":"Kresse","year":"2022"},{"key":"mlstac8fe0bib45","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.122.156001","article-title":"Low-scaling algorithm for nudged elastic band calculations using a surrogate machine learning model","volume":"122","author":"Garrido Torres","year":"2019","journal-title":"Phys. Rev. Lett."},{"key":"mlstac8fe0bib46","doi-asserted-by":"publisher","DOI":"10.1063\/1.4960708","article-title":"Acceleration of saddle-point searches with machine learning","volume":"145","author":"Peterson","year":"2016","journal-title":"J. Chem. Phys."},{"key":"mlstac8fe0bib47","first-page":"pp 4944","article-title":"Parameter-efficient transfer learning for NLP","volume":"vol 2019","author":"Houlsby","year":"2019"},{"key":"mlstac8fe0bib48","first-page":"pp 595","article-title":"Learning how to active learn: A deep reinforcement learning approach","author":"Fang","year":"2017"}],"container-title":["Machine Learning: Science and Technology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac8fe0","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac8fe0\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac8fe0","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac8fe0\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac8fe0\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac8fe0\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac8fe0\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac8fe0\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,9,20]],"date-time":"2022-09-20T10:13:30Z","timestamp":1663668810000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac8fe0"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,9,1]]},"references-count":48,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2022,9,20]]},"published-print":{"date-parts":[[2022,9,1]]}},"URL":"https:\/\/doi.org\/10.1088\/2632-2153\/ac8fe0","relation":{},"ISSN":["2632-2153"],"issn-type":[{"value":"2632-2153","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,9,1]]},"assertion":[{"value":"FINETUNA: fine-tuning accelerated molecular simulations","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":"2022-04-13","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2022-09-06","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2022-09-20","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}