{"id":"https://openalex.org/W7165659700","doi":"https://doi.org/10.48550/arxiv.2606.23357","title":"SOAP-Bubbles: Structured Weight Uncertainty for Neural Networks","display_name":"SOAP-Bubbles: Structured Weight Uncertainty for Neural Networks","publication_year":2026,"publication_date":"2026-06-22","ids":{"openalex":"https://openalex.org/W7165659700","doi":"https://doi.org/10.48550/arxiv.2606.23357"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.23357","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.23357","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2606.23357","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5116305979","display_name":"Adrian Robert Minut","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Minut, Adrian Robert","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075130141","display_name":"Nico Daheim","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Daheim, Nico","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5007280203","display_name":"Marco Miani","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Miani, Marco","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5053565482","display_name":"Mohammad Emtiyaz Khan","orcid":"https://orcid.org/0000-0001-7614-0489"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Khan, Mohammad Emtiyaz","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5123872746","display_name":"Wu Lin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lin, Wu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5008471432","display_name":"Thomas M\u00f6llenhoff","orcid":"https://orcid.org/0000-0001-7730-0843"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"M\u00f6llenhoff, Thomas","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.32010000944137573,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.32010000944137573,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.09989999979734421,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.08889999985694885,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/preconditioner","display_name":"Preconditioner","score":0.6844000220298767},{"id":"https://openalex.org/keywords/diagonal","display_name":"Diagonal","score":0.5929999947547913},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.5716999769210815},{"id":"https://openalex.org/keywords/covariance","display_name":"Covariance","score":0.5047000050544739},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.48579999804496765},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.4702000021934509},{"id":"https://openalex.org/keywords/gaussian","display_name":"Gaussian","score":0.4174000024795532}],"concepts":[{"id":"https://openalex.org/C167431342","wikidata":"https://www.wikidata.org/wiki/Q1754327","display_name":"Preconditioner","level":3,"score":0.6844000220298767},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6104999780654907},{"id":"https://openalex.org/C130367717","wikidata":"https://www.wikidata.org/wiki/Q189791","display_name":"Diagonal","level":2,"score":0.5929999947547913},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.5716999769210815},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.521399974822998},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5210999846458435},{"id":"https://openalex.org/C178650346","wikidata":"https://www.wikidata.org/wiki/Q201984","display_name":"Covariance","level":2,"score":0.5047000050544739},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.48579999804496765},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.4702000021934509},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.4318999946117401},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.4174000024795532},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3716999888420105},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.3693999946117401},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.34450000524520874},{"id":"https://openalex.org/C158693339","wikidata":"https://www.wikidata.org/wiki/Q190524","display_name":"Eigenvalues and eigenvectors","level":2,"score":0.3230000138282776},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3206000030040741},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.3158000111579895},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.31130000948905945},{"id":"https://openalex.org/C61326573","wikidata":"https://www.wikidata.org/wiki/Q1496376","display_name":"Gaussian process","level":3,"score":0.29840001463890076},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2928999960422516}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.23357","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.23357","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.23357","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.23357","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Structured":[0],"weight-uncertainty":[1],"can":[2,25],"improve":[3],"many":[4],"aspects":[5],"of":[6,48,73],"deep":[7,137],"learning,":[8],"but":[9],"it":[10,116,128],"remains":[11],"costly":[12],"to":[13,17,37,56,71,80,130],"estimate":[14,60,131],"and":[15,51,75,92,110],"difficult":[16],"implement.":[18],"Here,":[19],"we":[20],"show":[21,99],"that":[22],"these":[23],"issues":[24],"be":[26],"addressed":[27],"by":[28],"adapting":[29],"the":[30,46,54,58,85,106],"SOAP":[31,74],"optimizer.":[32],"Our":[33,125],"key":[34],"idea":[35],"is":[36],"run":[38],"IVON,":[39],"an":[40],"existing":[41,122],"diagonal-covariance":[42,123],"variational":[43],"method,":[44],"in":[45,88],"eigenspace":[47],"SOAP's":[49],"preconditioner":[50,55],"then":[52],"use":[53],"transform":[57],"diagonal":[59],"into":[61],"a":[62],"non-diagonal":[63],"covariance.":[64],"The":[65],"resulting":[66],"method":[67],"has":[68],"costs":[69],"similar":[70],"those":[72],"requires":[76],"no":[77],"drastic":[78],"changes":[79],"training":[81],"pipelines.":[82],"We":[83,98],"call":[84],"posteriors":[86],"obtained":[87],"this":[89],"way":[90],"SOAP-Bubbles":[91],"our":[93],"new":[94],"optimizer":[95],"Eigenspace-VON":[96],"(EVON).":[97],"that,":[100,111],"for":[101,112,136],"logistic":[102],"regression,":[103],"EVON":[104],"recovers":[105],"exact":[107],"Gaussian":[108],"covariance":[109],"language":[113],"model":[114],"pretraining,":[115],"yields":[117],"significantly":[118],"better":[119],"results":[120],"than":[121],"methods.":[124],"work":[126],"makes":[127],"easier":[129],"more":[132],"expressive":[133],"posterior":[134],"distributions":[135],"learning":[138],"at":[139],"scale.":[140]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-24T00:00:00"}