{"id":"https://openalex.org/W6924745688","doi":"https://doi.org/10.15496/publikation-76886","title":"Low-Cost Bayesian Methods for Fixing Neural Networks' Overconfidence","display_name":"Low-Cost Bayesian Methods for Fixing Neural Networks' Overconfidence","publication_year":2023,"publication_date":"2023-01-20","ids":{"openalex":"https://openalex.org/W6924745688","doi":"https://doi.org/10.15496/publikation-76886"},"language":"en","primary_location":{"id":"doi:10.15496/publikation-76886","is_oa":true,"landing_page_url":"https://doi.org/10.15496/publikation-76886","pdf_url":null,"source":{"id":"https://openalex.org/S7407053000","display_name":"Universit\u00e4tsbibliothek T\u00fcbingen","issn_l":null,"issn":[],"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"article"},"type":"other","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.15496/publikation-76886","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Kristiadi, Agustinus","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Kristiadi, Agustinus","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":1,"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":true,"primary_topic":null,"topics":[],"keywords":[{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.7907999753952026},{"id":"https://openalex.org/keywords/bayesian-probability","display_name":"Bayesian probability","score":0.6746000051498413},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.4724999964237213},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.4519999921321869},{"id":"https://openalex.org/keywords/overconfidence-effect","display_name":"Overconfidence effect","score":0.4406000077724457},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.4336000084877014},{"id":"https://openalex.org/keywords/parametric-statistics","display_name":"Parametric statistics","score":0.3912999927997589}],"concepts":[{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.7907999753952026},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6773999929428101},{"id":"https://openalex.org/C107673813","wikidata":"https://www.wikidata.org/wiki/Q812534","display_name":"Bayesian probability","level":2,"score":0.6746000051498413},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6485999822616577},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5644000172615051},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.4724999964237213},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.4519999921321869},{"id":"https://openalex.org/C51110983","wikidata":"https://www.wikidata.org/wiki/Q16503490","display_name":"Overconfidence effect","level":2,"score":0.4406000077724457},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.4336000084877014},{"id":"https://openalex.org/C117251300","wikidata":"https://www.wikidata.org/wiki/Q1849855","display_name":"Parametric statistics","level":2,"score":0.3912999927997589},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.3495999872684479},{"id":"https://openalex.org/C160234255","wikidata":"https://www.wikidata.org/wiki/Q812535","display_name":"Bayesian inference","level":3,"score":0.33149999380111694},{"id":"https://openalex.org/C33724603","wikidata":"https://www.wikidata.org/wiki/Q812540","display_name":"Bayesian network","level":2,"score":0.33059999346733093},{"id":"https://openalex.org/C97937538","wikidata":"https://www.wikidata.org/wiki/Q199691","display_name":"Laplace transform","level":2,"score":0.3190000057220459},{"id":"https://openalex.org/C71983512","wikidata":"https://www.wikidata.org/wiki/Q7915687","display_name":"Variable-order Bayesian network","level":4,"score":0.3061000108718872},{"id":"https://openalex.org/C22243797","wikidata":"https://www.wikidata.org/wiki/Q2058297","display_name":"Laplace's method","level":3,"score":0.302700012922287},{"id":"https://openalex.org/C207201462","wikidata":"https://www.wikidata.org/wiki/Q182505","display_name":"Bayes' theorem","level":3,"score":0.274399995803833},{"id":"https://openalex.org/C101112237","wikidata":"https://www.wikidata.org/wiki/Q4874481","display_name":"Bayesian statistics","level":4,"score":0.27379998564720154}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.15496/publikation-76886","is_oa":true,"landing_page_url":"https://doi.org/10.15496/publikation-76886","pdf_url":null,"source":{"id":"https://openalex.org/S7407053000","display_name":"Universit\u00e4tsbibliothek T\u00fcbingen","issn_l":null,"issn":[],"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"article"}],"best_oa_location":{"id":"doi:10.15496/publikation-76886","is_oa":true,"landing_page_url":"https://doi.org/10.15496/publikation-76886","pdf_url":null,"source":{"id":"https://openalex.org/S7407053000","display_name":"Universit\u00e4tsbibliothek T\u00fcbingen","issn_l":null,"issn":[],"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"article"},"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":{"Well-calibrated":[0],"predictive":[1],"uncertainty":[2,144],"of":[3,112,116,145,166],"neural":[4,20,41,87,119,135,187],"networks\u2014essentially":[5],"making":[6],"them":[7],"know":[8],"when":[9],"they":[10],"do":[11],"not":[12],"know\u2014is":[13],"paramount":[14],"in":[15,24,63,73,100,121,169,189],"safety-critical":[16],"applications.":[17],"However,":[18],"deep":[19,74,170,182],"networks":[21,42,88,120],"are":[22],"overconfident":[23,99],"the":[25,32,59,101,105,114,122,125,133,143,163],"region":[26,102],"both":[27],"far":[28,103],"away":[29],"and":[30,43,62,148],"near":[31,124],"training":[33,142],"data.":[34,106],"In":[35],"this":[36,48],"thesis,":[37],"we":[38,51,78,108],"study":[39],"Bayesian":[40,86,118,134,186],"their":[44],"extensions":[45],"to":[46,84,132,161,177],"mitigate":[47,71],"issue.":[49],"First,":[50],"show":[52],"that":[53,89,93],"being":[54],"Bayesian,":[55],"even":[56],"just":[57],"at":[58],"last":[60],"layer":[61],"a":[64,80,91,175,179,185,190],"post-hoc":[65],"manner":[66],"via":[67],"Laplace":[68,146,167],"approximations,":[69,147],"helps":[70],"overconfidence":[72],"ReLU":[75,85,94],"classifiers.":[76],"Then,":[77],"provide":[79,156],"cost-effective":[81],"Gaussian-process":[82],"extension":[83],"provides":[90],"guarantee":[92],"nets":[95],"will":[96],"never":[97],"be":[98],"from":[104],"Furthermore,":[107],"propose":[109],"three":[110],"ways":[111],"improving":[113],"calibration":[115],"general":[117],"regions":[123],"data":[126,152],"by":[127],"(i)":[128],"refining":[129],"parametric":[130],"approximations":[131,168],"networks\u2019":[136],"posteriors":[137],"with":[138],"normalizing":[139],"flows,":[140],"(ii)":[141],"(iii)":[149],"leveraging":[150],"out-of-distribution":[151],"during":[153],"training.":[154],"We":[155],"an":[157],"easy-to-use":[158],"library,":[159],"laplace-torch,":[160],"facilitate":[162],"modern":[164],"arts":[165],"learning.":[171],"It":[172],"gives":[173],"users":[174],"way":[176],"turn":[178],"standard":[180],"pre-trained":[181],"net":[183],"into":[184],"network":[188],"cost-efficient":[191],"manner.":[192]},"counts_by_year":[],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}