{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,9,8]],"date-time":"2025-09-08T06:01:57Z","timestamp":1757311317894,"version":"3.37.3"},"reference-count":41,"publisher":"IOP Publishing","issue":"1","license":[{"start":{"date-parts":[[2022,3,25]],"date-time":"2022-03-25T00:00:00Z","timestamp":1648166400000},"content-version":"vor","delay-in-days":24,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2022,3,25]],"date-time":"2022-03-25T00:00:00Z","timestamp":1648166400000},"content-version":"tdm","delay-in-days":24,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"DOI":"10.13039\/100000015","name":"U.S. Department of Energy","doi-asserted-by":"crossref","award":["DE-AC52-07NA27344"],"award-info":[{"award-number":["DE-AC52-07NA27344"]}],"id":[{"id":"10.13039\/100000015","id-type":"DOI","asserted-by":"crossref"}]},{"name":"LLNL-LDRD Program","award":["18-SI-002"],"award-info":[{"award-number":["18-SI-002"]}]}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Mach. Learn.: Sci. Technol."],"published-print":{"date-parts":[[2022,3,1]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>Many problems in science and engineering require making predictions based on few observations. To build a robust predictive model, these sparse data may need to be augmented with simulated data, especially when the design space is multi-dimensional. Simulations, however, often suffer from an inherent bias. Estimation of this bias may be poorly constrained not only because of data sparsity, but also because traditional predictive models fit only one type of observed outputs, such as scalars or images, instead of all available output data modalities, which might have been acquired and simulated at great cost. To break this limitation and open up the path for multi-modal calibration, we propose to combine a novel, transfer learning technique for suppressing the bias with recent developments in deep learning, which allow building predictive models with multi-modal outputs. First, we train an initial neural network model on simulated data to learn important correlations between different output modalities and between simulation inputs and outputs. Then, the model is partially retrained, or transfer learned, to fit the experiments; a method that has never been implemented in this type of architecture. Using fewer than 10 inertial confinement fusion experiments for training, transfer learning systematically improves the simulation predictions while a simple output calibration, which we design as a baseline, makes the predictions worse. We also offer extensive cross-validation with real and carefully designed synthetic data. The method described in this paper can be applied to a wide range of problems that require transferring knowledge from simulations to the domain of experiments.<\/jats:p>","DOI":"10.1088\/2632-2153\/ac5e3e","type":"journal-article","created":{"date-parts":[[2022,3,25]],"date-time":"2022-03-25T09:33:46Z","timestamp":1648200826000},"page":"015035","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":15,"title":["Suppressing simulation bias in multi-modal data using transfer learning"],"prefix":"10.1088","volume":"3","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-2085-3206","authenticated-orcid":true,"given":"Bogdan","family":"Kustowski","sequence":"first","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2408-0047","authenticated-orcid":false,"given":"Jim A","family":"Gaffney","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6946-8996","authenticated-orcid":false,"given":"Brian K","family":"Spears","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4169-5553","authenticated-orcid":true,"given":"Gemma J","family":"Anderson","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4186-3502","authenticated-orcid":false,"given":"Rushil","family":"Anirudh","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4107-3831","authenticated-orcid":false,"given":"Peer-Timo","family":"Bremer","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8517-5816","authenticated-orcid":false,"given":"Jayaraman J","family":"Thiagarajan","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8901-3857","authenticated-orcid":false,"given":"Michael K","family":"G Kruse","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5478-8820","authenticated-orcid":false,"given":"Ryan C","family":"Nora","sequence":"additional","affiliation":[]}],"member":"266","published-online":{"date-parts":[[2022,3,25]]},"reference":[{"key":"mlstac5e3ebib1","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/BF01197708","article-title":"A rigorous framework for optimization of expensive functions by surrogates","volume":"17","author":"Booker","year":"1999","journal-title":"Struct. Optim."},{"year":"2018","author":"Santner","key":"mlstac5e3ebib2"},{"key":"mlstac5e3ebib3","doi-asserted-by":"publisher","first-page":"425","DOI":"10.1111\/1467-9868.00294","article-title":"Bayesian calibration of computer models","volume":"63","author":"Kennedy","year":"2001","journal-title":"J. R. Stat. Soc. B"},{"key":"mlstac5e3ebib4","doi-asserted-by":"publisher","first-page":"1645","DOI":"10.1016\/j.future.2013.01.010","article-title":"Internet of Things (IoT): a vision, architectural elements and future directions","volume":"29","author":"Gubbia","year":"2013","journal-title":"Future Gener. Comput. Syst."},{"key":"mlstac5e3ebib5","doi-asserted-by":"publisher","first-page":"351","DOI":"10.1038\/s41586-021-03382-w","article-title":"The data-driven future of high-energy density physics","volume":"593","author":"Hatfield","year":"2021","journal-title":"Nature"},{"key":"mlstac5e3ebib6","doi-asserted-by":"publisher","first-page":"1247","DOI":"10.21105\/joss.01247","article-title":"SkyPortal: an astronomical data platform","volume":"4","author":"van der Walt","year":"2019","journal-title":"J. Open Source Softw."},{"key":"mlstac5e3ebib7","doi-asserted-by":"publisher","first-page":"KS149","DOI":"10.1190\/geo2019-0774.1","article-title":"Application of machine learning to microseismic event detection in distributed acoustic sensing data","volume":"85","author":"Stork","year":"2020","journal-title":"Geophysics"},{"key":"mlstac5e3ebib8","doi-asserted-by":"publisher","first-page":"1637","DOI":"10.1109\/JIOT.2017.2786639","article-title":"Internet of Things (IoT): research, simulators and testbeds","volume":"5","author":"Chernyshev","year":"2018","journal-title":"IEEE Internet Things J."},{"key":"mlstac5e3ebib9","doi-asserted-by":"publisher","first-page":"230","DOI":"10.1002\/sam.11344","article-title":"Ensemble simulations of inertial confinement fusion implosions","volume":"10","author":"Nora","year":"2017","journal-title":"Stat. Anal. Data Min."},{"key":"mlstac5e3ebib10","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1186\/s40668-019-0032-1","article-title":"Cosmological N-body simulations: a challenge for scalable generative models","volume":"6","author":"Perraudin","year":"2019","journal-title":"Comput. Astrophys. Cosmol."},{"key":"mlstac5e3ebib11","doi-asserted-by":"publisher","first-page":"4755","DOI":"10.1109\/JLT.2019.2919713","article-title":"Efficient processing of distributed acoustic sensing data using a deep learning approach","volume":"37","author":"Shiloh","year":"2019","journal-title":"J. Lightwave Technol."},{"key":"mlstac5e3ebib12","first-page":"pp 656","article-title":"Learning many related tasks at the same time with backpropagation","volume":"vol 7","author":"Caruna","year":"1995"},{"key":"mlstac5e3ebib13","first-page":"pp 3320","article-title":"How transferable are features in deep neural networks?","volume":"vol 27","author":"Yosinski","year":"2014"},{"key":"mlstac5e3ebib14","doi-asserted-by":"publisher","first-page":"1345","DOI":"10.1109\/TKDE.2009.191","article-title":"A survey on transfer learning","volume":"22","author":"Pan","year":"2010","journal-title":"IEEE Trans. Knowl. Data Eng."},{"key":"mlstac5e3ebib15","first-page":"pp 608","article-title":"Mapping and revising Markov logic networks for transfer learning","author":"Mihalkova","year":"2007"},{"key":"mlstac5e3ebib16","first-page":"691","article-title":"A case study on meta-generalising: a Gaussian processes approach","volume":"13","author":"Skolidis","year":"2012","journal-title":"J. Mach. Learn. Res."},{"key":"mlstac5e3ebib17","doi-asserted-by":"publisher","first-page":"46","DOI":"10.1109\/TPS.2019.2948339","article-title":"Transfer learning as a tool for reducing simulation bias: application to inertial confinement fusion","volume":"48","author":"Kustowski","year":"2020","journal-title":"IEEE Trans. Plasma Sci."},{"key":"mlstac5e3ebib18","doi-asserted-by":"publisher","first-page":"61","DOI":"10.1109\/TPS.2019.2955098","article-title":"Transfer learning to model inertial confinement fusion experiments","volume":"48","author":"Humbird","year":"2020","journal-title":"IEEE Trans. Plasma Sci."},{"key":"mlstac5e3ebib19","first-page":"pp 185","article-title":"Transfer-learning: bridging the gap between real and simulation data for machine learning in injection modeling","volume":"vol 72","author":"Tercan","year":"2018"},{"key":"mlstac5e3ebib20","doi-asserted-by":"publisher","first-page":"9741","DOI":"10.1073\/pnas.1916634117","article-title":"Improved surrogates in inertial confinement fusion with manifold and cycle consistencies","volume":"117","author":"Anirudh","year":"2020","journal-title":"Proc. Natl Acad. Sci."},{"article-title":"Generalizing from a few examples: a survey on few-shot learning","year":"2020","author":"Wang","key":"mlstac5e3ebib21"},{"year":"2009","author":"Atzeni","key":"mlstac5e3ebib22"},{"key":"mlstac5e3ebib23","doi-asserted-by":"publisher","first-page":"435","DOI":"10.1038\/nphys3736","article-title":"Inertial-confinement fusion with lasers","volume":"12","author":"Betti","year":"2016","journal-title":"Nat. Phys."},{"key":"mlstac5e3ebib24","doi-asserted-by":"publisher","first-page":"S228","DOI":"10.1088\/0029-5515\/44\/12\/S14","article-title":"The National Ignition Facility: enabling fusion ignition for the 21st century","volume":"44","author":"Miller","year":"2004","journal-title":"Nucl. Fusion"},{"key":"mlstac5e3ebib25","doi-asserted-by":"publisher","DOI":"10.1063\/1.4977912","article-title":"Zonal flow generation in inertial confinement fusion implosions","volume":"24","author":"Peterson","year":"2017","journal-title":"Phys. Plasmas"},{"key":"mlstac5e3ebib26","doi-asserted-by":"publisher","DOI":"10.1063\/1.5091985","article-title":"The blind implosion-maker: automated inertial confinement fusion experiment design","volume":"26","author":"Hatfield","year":"2019","journal-title":"Phys. Plasmas"},{"key":"mlstac5e3ebib27","doi-asserted-by":"publisher","DOI":"10.1063\/1.5019741","article-title":"The high velocity, high adiabat, \u201cBigfoot\u201d campaign and tests of indirect-drive implosion scaling","volume":"25","author":"Casey","year":"2018","journal-title":"Phys. Plasmas"},{"key":"mlstac5e3ebib28","doi-asserted-by":"publisher","DOI":"10.1063\/1.5108667","article-title":"Making inertial confinement fusion models more predictive","volume":"26","author":"Gaffney","year":"2019","journal-title":"Phys. Plasmas"},{"key":"mlstac5e3ebib29","doi-asserted-by":"publisher","first-page":"2275","DOI":"10.1063\/1.1356740","article-title":"Three-dimensional HYDRA simulations of National Ignition Facility targets","volume":"8","author":"Marinak","year":"2001","journal-title":"Phys. Plasmas"},{"key":"mlstac5e3ebib30","doi-asserted-by":"publisher","DOI":"10.1063\/1.4906897","article-title":"Radiation hydrodynamics modeling of the highest compression inertial confinement fusion ignition experiment from the national ignition campaign","volume":"22","author":"Clark","year":"2015","journal-title":"Phys. Plasmas"},{"article-title":"Early predictions of the simulation outputs","year":"2019","author":"Kustowski","key":"mlstac5e3ebib31"},{"article-title":"Wasserstein auto-encoders","year":"2018","author":"Tolstikhin","key":"mlstac5e3ebib32"},{"key":"mlstac5e3ebib33","doi-asserted-by":"crossref","DOI":"10.1609\/aaai.v32i1.11651","article-title":"Measuring catastrophic forgetting in neural networks","author":"Kemker","year":"2018"},{"key":"mlstac5e3ebib34","doi-asserted-by":"publisher","first-page":"873","DOI":"10.1364\/JOSAA.21.000873","article-title":"Ince\u2013Gaussian modes of the paraxial wave equation and stable resonators","volume":"21","author":"Bandres","year":"2004","journal-title":"J. Opt. Soc. Am. A"},{"article-title":"Area-based image metrics elucidate differences between radiation-hydrodynamics simulations and NIF experimental x-ray images","year":"2019","author":"Kruse","key":"mlstac5e3ebib35"},{"article-title":"Rapid thermal and data-driven models for stagnating ICF implosions","year":"2015","author":"Gaffney","key":"mlstac5e3ebib36"},{"key":"mlstac5e3ebib37","doi-asserted-by":"publisher","DOI":"10.1051\/epjconf\/20135904001","article-title":"Integrated thermodynamic model for ignition target performance","volume":"59","author":"Springer","year":"2013","journal-title":"EPJ Web Conf."},{"key":"mlstac5e3ebib38","doi-asserted-by":"publisher","first-page":"581","DOI":"10.1038\/s41586-019-0877-0","article-title":"Tripled yield in direct-drive laser fusion through statistical modelling","volume":"565","author":"Gopalaswamy","year":"2019","journal-title":"Nature"},{"key":"mlstac5e3ebib39","doi-asserted-by":"publisher","DOI":"10.1063\/5.0041907","article-title":"Cognitive simulation models for inertial confinement fusion: combining simulation and experimental data","volume":"28","author":"Humbird","year":"2021","journal-title":"Phys. Plasmas"},{"author":"Anderson","key":"mlstac5e3ebib40","article-title":"Meaningful uncertainties from deep neuralnetwork surrogates of large-scale numericalsimulations"},{"key":"mlstac5e3ebib41","doi-asserted-by":"crossref","DOI":"10.1007\/s11263-020-01310-5","article-title":"MimicGAN: robust projection onto image manifolds with corruption mimicking","author":"Anirudh","year":"2020"}],"container-title":["Machine Learning: Science and Technology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac5e3e","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac5e3e\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac5e3e\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac5e3e\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,1,30]],"date-time":"2023-01-30T02:29:34Z","timestamp":1675045774000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ac5e3e"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,3,1]]},"references-count":41,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2022,3,25]]},"published-print":{"date-parts":[[2022,3,1]]}},"URL":"https:\/\/doi.org\/10.1088\/2632-2153\/ac5e3e","relation":{},"ISSN":["2632-2153"],"issn-type":[{"type":"electronic","value":"2632-2153"}],"subject":[],"published":{"date-parts":[[2022,3,1]]},"assertion":[{"value":"Suppressing simulation bias in multi-modal data using transfer learning","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":"2021-12-17","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2022-03-16","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2022-03-25","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}