{"id":"https://openalex.org/W4403488038","doi":"https://doi.org/10.3233/faia240658","title":"Improving Calibration by Relating Focal Loss, Temperature Scaling, and Properness","display_name":"Improving Calibration by Relating Focal Loss, Temperature Scaling, and Properness","publication_year":2024,"publication_date":"2024-10-16","ids":{"openalex":"https://openalex.org/W4403488038","doi":"https://doi.org/10.3233/faia240658"},"language":"en","primary_location":{"id":"doi:10.3233/faia240658","is_oa":true,"landing_page_url":"https://doi.org/10.3233/faia240658","pdf_url":"https://ebooks.iospress.nl/pdf/doi/10.3233/FAIA240658","source":{"id":"https://openalex.org/S4210201731","display_name":"Frontiers in artificial intelligence and applications","issn_l":"0922-6389","issn":["0922-6389","1879-8314"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":"cc-by-nc","license_id":"https://openalex.org/licenses/cc-by-nc","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Frontiers in Artificial Intelligence and Applications","raw_type":"book-chapter"},"type":"book-chapter","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://ebooks.iospress.nl/pdf/doi/10.3233/FAIA240658","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5075282377","display_name":"Viacheslav Komisarenko","orcid":"https://orcid.org/0000-0001-9210-1260"},"institutions":[{"id":"https://openalex.org/I56085075","display_name":"University of Tartu","ror":"https://ror.org/03z77qz90","country_code":"EE","type":"education","lineage":["https://openalex.org/I56085075"]}],"countries":["EE"],"is_corresponding":false,"raw_author_name":"Viacheslav Komisarenko","raw_affiliation_strings":["Institute of Computer Science, University of Tartu, Estonia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Computer Science, University of Tartu, Estonia","institution_ids":["https://openalex.org/I56085075"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5087318731","display_name":"Meelis Kull","orcid":"https://orcid.org/0000-0001-9257-595X"},"institutions":[{"id":"https://openalex.org/I56085075","display_name":"University of Tartu","ror":"https://ror.org/03z77qz90","country_code":"EE","type":"education","lineage":["https://openalex.org/I56085075"]}],"countries":["EE"],"is_corresponding":false,"raw_author_name":"Meelis Kull","raw_affiliation_strings":["Institute of Computer Science, University of Tartu, Estonia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Computer Science, University of Tartu, Estonia","institution_ids":["https://openalex.org/I56085075"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I56085075"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.37262721,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"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":0.9455000162124634,"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":0.9455000162124634,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.021400000900030136,"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/T11652","display_name":"Imbalanced Data Classification Techniques","score":0.006899999920278788,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/scaling","display_name":"Scaling","score":0.6447973251342773},{"id":"https://openalex.org/keywords/calibration","display_name":"Calibration","score":0.5134969353675842},{"id":"https://openalex.org/keywords/environmental-science","display_name":"Environmental science","score":0.3324775993824005},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.28071728348731995},{"id":"https://openalex.org/keywords/statistics","display_name":"Statistics","score":0.13029921054840088},{"id":"https://openalex.org/keywords/geometry","display_name":"Geometry","score":0.06816750764846802}],"concepts":[{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.6447973251342773},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.5134969353675842},{"id":"https://openalex.org/C39432304","wikidata":"https://www.wikidata.org/wiki/Q188847","display_name":"Environmental science","level":0,"score":0.3324775993824005},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.28071728348731995},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.13029921054840088},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.06816750764846802}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.3233/faia240658","is_oa":true,"landing_page_url":"https://doi.org/10.3233/faia240658","pdf_url":"https://ebooks.iospress.nl/pdf/doi/10.3233/FAIA240658","source":{"id":"https://openalex.org/S4210201731","display_name":"Frontiers in artificial intelligence and applications","issn_l":"0922-6389","issn":["0922-6389","1879-8314"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":"cc-by-nc","license_id":"https://openalex.org/licenses/cc-by-nc","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Frontiers in Artificial Intelligence and Applications","raw_type":"book-chapter"},{"id":"pmh:oai:arXiv.org:2408.11598","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2408.11598","pdf_url":"https://arxiv.org/pdf/2408.11598","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"doi:10.3233/faia240658","is_oa":true,"landing_page_url":"https://doi.org/10.3233/faia240658","pdf_url":"https://ebooks.iospress.nl/pdf/doi/10.3233/FAIA240658","source":{"id":"https://openalex.org/S4210201731","display_name":"Frontiers in artificial intelligence and applications","issn_l":"0922-6389","issn":["0922-6389","1879-8314"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":"cc-by-nc","license_id":"https://openalex.org/licenses/cc-by-nc","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Frontiers in Artificial Intelligence and Applications","raw_type":"book-chapter"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1653152820","display_name":"Contextual uncertainty and representation learning in machine perception","funder_award_id":"PRG1604","funder_id":"https://openalex.org/F4320321090","funder_display_name":"Eesti Teadusagentuur"}],"funders":[{"id":"https://openalex.org/F4320321090","display_name":"Eesti Teadusagentuur","ror":"https://ror.org/00jjeja18"}],"has_content":{"pdf":true,"grobid_xml":false},"content_urls":{"pdf":"https://content.openalex.org/works/W4403488038.pdf"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W141820298","https://openalex.org/W2049584446","https://openalex.org/W2079781215","https://openalex.org/W4378770497","https://openalex.org/W4308245303","https://openalex.org/W2014033564","https://openalex.org/W2910573937","https://openalex.org/W4385571583","https://openalex.org/W4389519396","https://openalex.org/W2113438243"],"abstract_inverted_index":{"Proper":[0],"losses":[1],"such":[2,36],"as":[3,37,153],"cross-entropy":[4,83],"incentivize":[5],"classifiers":[6,24,57],"to":[7,19,26,53,79,111,129,152],"produce":[8],"class":[9],"probabilities":[10],"that":[11,58,89,184],"are":[12,59],"well-calibrated":[13],"on":[14,29,62,115,124,178],"the":[15,20,30,109,116,125,130,154],"training":[16,47,76,117],"data.":[17,64],"Due":[18],"generalization":[21,131],"gap,":[22],"these":[23],"tend":[25],"become":[27],"overconfident":[28],"test":[31,63,126],"data,":[32,118,127],"mandating":[33],"calibration":[34,81,156,168,172],"methods":[35],"temperature":[38,140,162,174,186,190],"scaling.":[39,175,191],"The":[40],"focal":[41,74,90,106,143,155,161,171,185],"loss":[42,75,91,107,144],"is":[43,68,104],"not":[44],"proper,":[45],"but":[46],"with":[48],"it":[49],"has":[50],"been":[51],"shown":[52],"often":[54,77],"result":[55],"in":[56,120],"better":[60,80,122],"calibrated":[61,123],"Our":[65,176],"first":[66],"contribution":[67],"a":[69,96,100,136,165],"simple":[70],"explanation":[71],"about":[72],"why":[73,105],"leads":[78],"than":[82],"training.":[84],"For":[85],"this,":[86],"we":[87,134,150,159],"prove":[88],"can":[92],"be":[93],"decomposed":[94],"into":[95],"confidence-raising":[97,147],"transformation":[98],"and":[99,142,173],"proper":[101],"loss.":[102],"This":[103],"pushes":[108],"model":[110],"provide":[112],"under-confident":[113],"predictions":[114],"resulting":[119],"being":[121],"due":[128],"gap.":[132],"Secondly,":[133],"reveal":[135],"strong":[137],"connection":[138],"between":[139],"scaling":[141,163,187],"through":[145],"its":[146],"transformation,":[148],"which":[149],"refer":[151],"map.":[157],"Thirdly,":[158],"propose":[160],"-":[164],"new":[166],"post-hoc":[167],"method":[169],"combining":[170],"experiments":[177],"three":[179],"image":[180],"classification":[181],"datasets":[182],"demonstrate":[183],"outperforms":[188],"standard":[189]},"counts_by_year":[],"updated_date":"2026-06-26T08:34:08.712188","created_date":"2025-10-10T00:00:00"}