{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,21]],"date-time":"2026-03-21T07:07:25Z","timestamp":1774076845179,"version":"3.50.1"},"reference-count":51,"publisher":"IOP Publishing","issue":"1","license":[{"start":{"date-parts":[[2023,2,22]],"date-time":"2023-02-22T00:00:00Z","timestamp":1677024000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2023,2,22]],"date-time":"2023-02-22T00:00:00Z","timestamp":1677024000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"DOI":"10.13039\/501100000271","name":"Science and Technology Facilities Council","doi-asserted-by":"crossref","award":["ST\/T000473\/1"],"award-info":[{"award-number":["ST\/T000473\/1"]}],"id":[{"id":"10.13039\/501100000271","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":[[2023,3,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    Simulation-based inference (SBI) is rapidly establishing itself as a standard machine learning technique for analyzing data in cosmological surveys. Despite continual improvements to the quality of density estimation by learned models, applications of such techniques to real data are entirely reliant on the generalization power of neural networks far outside the training distribution, which is mostly unconstrained. Due to the imperfections in scientist-created simulations, and the large computational expense of generating all possible parameter combinations, SBI methods in cosmology are vulnerable to such generalization issues. Here, we discuss the effects of both issues, and show how using a Bayesian neural network framework for training SBI can mitigate biases, and result in more reliable inference outside the training set. We introduce\n                    <jats:monospace>cosmoSWAG<\/jats:monospace>\n                    , the first application of stochastic weight averaging to cosmology, and apply it to SBI trained for inference on the cosmic microwave background.\n                  <\/jats:p>","DOI":"10.1088\/2632-2153\/acbb53","type":"journal-article","created":{"date-parts":[[2023,2,10]],"date-time":"2023-02-10T17:38:48Z","timestamp":1676050728000},"page":"01LT01","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":22,"title":["Robust simulation-based inference in cosmology with Bayesian neural networks"],"prefix":"10.1088","volume":"4","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-4728-8473","authenticated-orcid":true,"given":"Pablo","family":"Lemos","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Miles","family":"Cranmer","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Muntazir","family":"Abidi","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"ChangHoon","family":"Hahn","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Michael","family":"Eickenberg","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Elena","family":"Massara","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"David","family":"Yallup","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shirley","family":"Ho","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"266","published-online":{"date-parts":[[2023,2,22]]},"reference":[{"key":"mlstacbb53bib1","doi-asserted-by":"publisher","DOI":"10.1016\/j.dark.2019.100369","volume":"26","author":"Kangal","year":"2019","journal-title":"Phys. Dark Universe"},{"key":"mlstacbb53bib2","author":"Ntampaka","year":"2019"},{"key":"mlstacbb53bib3","doi-asserted-by":"publisher","first-page":"008","DOI":"10.1088\/1475-7516\/2020\/03\/008","volume":"2020","author":"Escamilla-Rivera","year":"2020","journal-title":"J. Cosmol. Astropart. Phys."},{"key":"mlstacbb53bib4","doi-asserted-by":"publisher","DOI":"10.1016\/j.cpc.2020.107809","volume":"261","author":"Tilaver","year":"2021","journal-title":"Comput. Phys. Commun."},{"key":"mlstacbb53bib5","doi-asserted-by":"publisher","DOI":"10.1016\/j.ascom.2021.100504","volume":"37","author":"Salti","year":"2021","journal-title":"Astron. Comput."},{"key":"mlstacbb53bib6","author":"Dvorkin","year":"2022"},{"key":"mlstacbb53bib7","doi-asserted-by":"publisher","first-page":"410","DOI":"10.1016\/j.tree.2010.04.001","volume":"25","author":"Csill\u00e9ry","year":"2010","journal-title":"Trends Ecol. Evol."},{"key":"mlstacbb53bib8","doi-asserted-by":"publisher","first-page":"379","DOI":"10.1146\/annurev-ecolsys-102209-144621","volume":"41","author":"Beaumont","year":"2010","journal-title":"Annu. Rev. Ecol. Evol. Syst."},{"key":"mlstacbb53bib9","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pcbi.1002803","volume":"9","author":"Sunn\u00e5ker","year":"2013","journal-title":"PLoS Comput. Biol."},{"key":"mlstacbb53bib10","author":"Thomas","year":"2016"},{"key":"mlstacbb53bib11","doi-asserted-by":"publisher","first-page":"L85","DOI":"10.1093\/mnrasl\/slab081","volume":"506","author":"Leclercq","year":"2021","journal-title":"Mon. Not. R. Astron. Soc."},{"key":"mlstacbb53bib12","doi-asserted-by":"publisher","first-page":"1","DOI":"10.2202\/1544-6115.1684","volume":"10","author":"Bonassi","year":"2011","journal-title":"Stat. Appl. Genet. Mol. Biol."},{"key":"mlstacbb53bib13","doi-asserted-by":"publisher","first-page":"34","DOI":"10.1002\/sta4.15","volume":"2","author":"Fan","year":"2013","journal-title":"Stat"},{"key":"mlstacbb53bib14","volume":"vol 29","author":"Papamakarios","year":"2016"},{"key":"mlstacbb53bib15","volume":"vol 30","author":"Lueckmann","year":"2017"},{"key":"mlstacbb53bib16","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.103.023009","volume":"103","author":"Lemos","year":"2021","journal-title":"Phys. Rev. D"},{"key":"mlstacbb53bib17","doi-asserted-by":"publisher","first-page":"30055","DOI":"10.1073\/pnas.1912789117","volume":"117","author":"Cranmer","year":"2020","journal-title":"Proc. Natl Acad. Sci."},{"key":"mlstacbb53bib18","doi-asserted-by":"publisher","first-page":"4244","DOI":"10.1093\/mnras\/stx2326","volume":"472","author":"Heavens","year":"2017","journal-title":"Mon. Not. R. Astron. Soc."},{"key":"mlstacbb53bib19","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.97.083004","volume":"97","author":"Charnock","year":"2018","journal-title":"Phys. Rev. D"},{"key":"mlstacbb53bib20","doi-asserted-by":"publisher","first-page":"049","DOI":"10.1088\/1475-7516\/2021\/11\/049","volume":"2021","author":"Makinen","year":"2021","journal-title":"J. Cosmol. Astropart. Phys."},{"key":"mlstacbb53bib21","doi-asserted-by":"publisher","first-page":"2","DOI":"10.3847\/1538-4365\/ab9d82","volume":"250","author":"Villaescusa-Navarro","year":"2020","journal-title":"Astrophys. J. Suppl. Ser."},{"key":"mlstacbb53bib22","doi-asserted-by":"publisher","first-page":"71","DOI":"10.3847\/1538-4357\/abf7ba","volume":"915","author":"Villaescusa-Navarro","year":"2021","journal-title":"Astrophys. J."},{"key":"mlstacbb53bib23","doi-asserted-by":"publisher","first-page":"361","DOI":"10.1007\/BF00200801","volume":"61","author":"Kononenko","year":"1989","journal-title":"Biol. Cybern."},{"key":"mlstacbb53bib24","doi-asserted-by":"publisher","first-page":"73","DOI":"10.1016\/0168-9002(94)00931-7","volume":"354","author":"MacKay","year":"1995","journal-title":"Nucl. Instrum. Methods Phys. Res. A"},{"key":"mlstacbb53bib25","first-page":"pp 1050","article-title":"Dropout as a bayesian approximation: representing model uncertainty in deep learning","author":"Gal","year":"2016"},{"key":"mlstacbb53bib26","author":"Yallup","year":"2022"},{"key":"mlstacbb53bib27","volume":"vol 32","author":"Maddox","year":"2019"},{"key":"mlstacbb53bib28","first-page":"pp 4697","volume":"vol 33","author":"Wilson","year":"2020"},{"key":"mlstacbb53bib29","doi-asserted-by":"publisher","first-page":"1091","DOI":"10.1073\/pnas.2026053118","volume":"118","author":"Cranmer","year":"2021","journal-title":"Proc. Natl Acad. Sci."},{"key":"mlstacbb53bib30","volume":"vol 3","author":"Gal","year":"2017"},{"key":"mlstacbb53bib31","volume":"vol 24","author":"Graves","year":"2011"},{"key":"mlstacbb53bib32","author":"Kingma","year":"2013"},{"key":"mlstacbb53bib33","author":"Cole","year":"2021"},{"key":"mlstacbb53bib34","doi-asserted-by":"publisher","first-page":"473","DOI":"10.1086\/309179","volume":"538","author":"Lewis","year":"2000","journal-title":"Astrophys. J."},{"key":"mlstacbb53bib35","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.66.103511","volume":"66","author":"Lewis","year":"2002","journal-title":"Phys. Rev. D"},{"key":"mlstacbb53bib36","doi-asserted-by":"publisher","first-page":"027","DOI":"10.1088\/1475-7516\/2012\/04\/027","volume":"2012","author":"Howlett","year":"2012","journal-title":"J. Cosmol. Astropart. Phys."},{"key":"mlstacbb53bib37","doi-asserted-by":"publisher","first-page":"A6","DOI":"10.1051\/0004-6361\/201833910e","volume":"641","author":"Aghanim","year":"2020","journal-title":"Astron. Astrophys."},{"key":"mlstacbb53bib38","doi-asserted-by":"publisher","first-page":"057","DOI":"10.1088\/1475-7516\/2021\/05\/057","volume":"2021","author":"Torrado","year":"2021","journal-title":"J. Cosmol. Astropart. Phys."},{"key":"mlstacbb53bib39","author":"Lewis","year":"2019"},{"key":"mlstacbb53bib40","doi-asserted-by":"publisher","first-page":"2874","DOI":"10.1093\/mnras\/sty819","volume":"477","author":"Alsing","year":"2018","journal-title":"Mon. Not. R. Astron. Soc."},{"key":"mlstacbb53bib41","doi-asserted-by":"publisher","first-page":"4440","DOI":"10.1093\/mnras\/stz1900","volume":"488","author":"Alsing","year":"2019","journal-title":"Mon. Not. R. Astron. Soc."},{"key":"mlstacbb53bib42","author":"Agarap","year":"2018"},{"key":"mlstacbb53bib43","volume":"vol 30","author":"Papamakarios","year":"2017"},{"key":"mlstacbb53bib44","first-page":"pp 881","article-title":"Made: masked autoencoder for distribution estimation","author":"Germain","year":"2015"},{"key":"mlstacbb53bib45","author":"Helminger","year":"2020"},{"key":"mlstacbb53bib46","author":"Bishop","year":"1994"},{"key":"mlstacbb53bib47","author":"Mandt","year":"2017"},{"key":"mlstacbb53bib48","first-page":"pp 1861","article-title":"Probabilistic backpropagation for scalable learning of bayesian neural networks","author":"Hern\u00e1ndez-Lobato","year":"2015"},{"key":"mlstacbb53bib49","doi-asserted-by":"publisher","first-page":"37","DOI":"10.1038\/386037a0","volume":"386","author":"Hu","year":"1997","journal-title":"Nature"},{"key":"mlstacbb53bib50","doi-asserted-by":"publisher","first-page":"L7","DOI":"10.3847\/2041-8213\/aa9704","volume":"850","author":"Levasseur","year":"2017","journal-title":"Astrophys. J. Lett."},{"key":"mlstacbb53bib51","author":"Hermans","year":"2021"}],"container-title":["Machine Learning: Science and Technology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acbb53","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acbb53\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acbb53","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acbb53\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acbb53\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acbb53\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acbb53\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acbb53\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,2,22]],"date-time":"2023-02-22T05:36:18Z","timestamp":1677044178000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acbb53"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,2,22]]},"references-count":51,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2023,2,22]]},"published-print":{"date-parts":[[2023,3,1]]}},"URL":"https:\/\/doi.org\/10.1088\/2632-2153\/acbb53","relation":{"has-review":[{"id-type":"doi","id":"10.1088\/2632-2153\/ACBB53\/v2\/decision1","asserted-by":"object"},{"id-type":"doi","id":"10.1088\/2632-2153\/ACBB53\/v2\/review2","asserted-by":"object"},{"id-type":"doi","id":"10.1088\/2632-2153\/ACBB53\/v1\/review2","asserted-by":"object"},{"id-type":"doi","id":"10.1088\/2632-2153\/ACBB53\/v1\/review1","asserted-by":"object"},{"id-type":"doi","id":"10.1088\/2632-2153\/ACBB53\/v2\/review1","asserted-by":"object"},{"id-type":"doi","id":"10.1088\/2632-2153\/ACBB53\/v2\/response1","asserted-by":"object"},{"id-type":"doi","id":"10.1088\/2632-2153\/ACBB53\/v3\/decision1","asserted-by":"object"},{"id-type":"doi","id":"10.1088\/2632-2153\/ACBB53\/v1\/decision1","asserted-by":"object"},{"id-type":"doi","id":"10.1088\/2632-2153\/ACBB53\/v3\/response1","asserted-by":"object"}]},"ISSN":["2632-2153"],"issn-type":[{"value":"2632-2153","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,2,22]]},"assertion":[{"value":"Robust simulation-based inference in cosmology with Bayesian 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 2023 The Author(s). Published by IOP Publishing Ltd","name":"copyright_information","label":"Copyright Information"},{"value":"2022-08-17","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2023-02-10","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2023-02-22","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}