{"id":"https://openalex.org/W2970565181","doi":"https://doi.org/10.48550/arxiv.1910.04851","title":"Addressing Failure Prediction by Learning Model Confidence","display_name":"Addressing Failure Prediction by Learning Model Confidence","publication_year":2019,"publication_date":"2019-10-01","ids":{"openalex":"https://openalex.org/W2970565181","doi":"https://doi.org/10.48550/arxiv.1910.04851","mag":"2970565181"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:1910.04851","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/1910.04851","pdf_url":"https://arxiv.org/pdf/1910.04851","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/1910.04851","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Corbi\u00e8re, Charles","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4210145724","display_name":"Centre d'\u00e9tudes et de recherche en informatique et communications","ror":"https://ror.org/044j5mm75","country_code":"FR","type":"facility","lineage":["https://openalex.org/I124158823","https://openalex.org/I4210134562","https://openalex.org/I4210145724"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Corbi\u00e8re, Charles","raw_affiliation_strings":["CEDRIC. M\u00e9thodes statistiques de data-mining et apprentissage","Valeo.ai"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"CEDRIC. M\u00e9thodes statistiques de data-mining et apprentissage","institution_ids":["https://openalex.org/I4210145724"]},{"raw_affiliation_string":"Valeo.ai","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5017490804","display_name":"Nicolas Thome","orcid":"https://orcid.org/0000-0003-4871-3045","observed_orcids":["https://orcid.org/0000-0003-4871-3045"]},"institutions":[{"id":"https://openalex.org/I124158823","display_name":"Conservatoire National des Arts et M\u00e9tiers","ror":"https://ror.org/0175hh227","country_code":"FR","type":"education","lineage":["https://openalex.org/I124158823","https://openalex.org/I4210134562"]},{"id":"https://openalex.org/I4210145724","display_name":"Centre d'\u00e9tudes et de recherche en informatique et communications","ror":"https://ror.org/044j5mm75","country_code":"FR","type":"facility","lineage":["https://openalex.org/I124158823","https://openalex.org/I4210134562","https://openalex.org/I4210145724"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Thome, Nicolas","raw_affiliation_strings":["CEDRIC , Conservatoire National des Arts et M\u00e9tiers , Paris , France","CEDRIC. M\u00e9thodes statistiques de data-mining et apprentissage","Machine Learning and Information Access"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"CEDRIC , Conservatoire National des Arts et M\u00e9tiers , Paris , France","institution_ids":["https://openalex.org/I124158823","https://openalex.org/I4210145724"]},{"raw_affiliation_string":"CEDRIC. M\u00e9thodes statistiques de data-mining et apprentissage","institution_ids":["https://openalex.org/I4210145724"]},{"raw_affiliation_string":"Machine Learning and Information Access","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015118459","display_name":"Avner Bar\u2010Hen","orcid":"https://orcid.org/0000-0002-4449-8117","observed_orcids":["https://orcid.org/0000-0002-4449-8117"]},"institutions":[{"id":"https://openalex.org/I4210145724","display_name":"Centre d'\u00e9tudes et de recherche en informatique et communications","ror":"https://ror.org/044j5mm75","country_code":"FR","type":"facility","lineage":["https://openalex.org/I124158823","https://openalex.org/I4210134562","https://openalex.org/I4210145724"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Bar-Hen, Avner","raw_affiliation_strings":["CEDRIC. M\u00e9thodes statistiques de data-mining et apprentissage"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"CEDRIC. M\u00e9thodes statistiques de data-mining et apprentissage","institution_ids":["https://openalex.org/I4210145724"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5108118084","display_name":"Matthieu Cord","orcid":"https://orcid.org/0000-0002-0627-5844","observed_orcids":["https://orcid.org/0000-0002-0627-5844"]},"institutions":[{"id":"https://openalex.org/I220619192","display_name":"Valeo (France)","ror":"https://ror.org/04ryqpf83","country_code":"FR","type":"company","lineage":["https://openalex.org/I220619192"]},{"id":"https://openalex.org/I39804081","display_name":"Sorbonne Universit\u00e9","ror":"https://ror.org/02en5vm52","country_code":"FR","type":"education","lineage":["https://openalex.org/I39804081"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Cord, Matthieu","raw_affiliation_strings":["Sorbonne University , Paris , France","valeo.ai , Paris , France","Machine Learning and Information Access"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Sorbonne University , Paris , France","institution_ids":["https://openalex.org/I39804081"]},{"raw_affiliation_string":"valeo.ai , Paris , France","institution_ids":["https://openalex.org/I220619192"]},{"raw_affiliation_string":"Machine Learning and Information Access","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5113908330","display_name":"Patrick P\u00e9rez","orcid":"https://orcid.org/0000-0002-8124-1206","observed_orcids":["https://orcid.org/0000-0002-8124-1206"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"P\u00e9rez, Patrick","raw_affiliation_strings":["Valeo.ai"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Valeo.ai","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":211,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"2898","last_page":"2909"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.8442000150680542,"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":0.8442000150680542,"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":0.12080000340938568,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.007400000002235174,"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/failure-prediction","display_name":"failure prediction","score":0.8889999985694885},{"id":"https://openalex.org/keywords/model-confidence","display_name":"model confidence","score":0.8790000081062317},{"id":"https://openalex.org/keywords/deep-neural-network","display_name":"deep neural network","score":0.7829999923706055},{"id":"https://openalex.org/keywords/confidence-calibration","display_name":"confidence calibration","score":0.5860000252723694},{"id":"https://openalex.org/keywords/uncertainty-estimation","display_name":"uncertainty estimation","score":0.6650000214576721},{"id":"https://openalex.org/keywords/bayesian-uncertainty","display_name":"Bayesian uncertainty","score":0.7580000162124634},{"id":"https://openalex.org/keywords/image-classification","display_name":"image classification","score":0.9139999747276306},{"id":"https://openalex.org/keywords/semantic-segmentation","display_name":"semantic segmentation","score":0.824999988079071},{"id":"https://openalex.org/keywords/theoretical-guarantees","display_name":"theoretical guarantees","score":0.5740000009536743}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6676760315895081},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.45841050148010254},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4071570634841919}],"mesh":[],"locations_count":3,"locations":[{"id":"pmh:oai:arXiv.org:1910.04851","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/1910.04851","pdf_url":"https://arxiv.org/pdf/1910.04851","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"pmh:oai:HAL:hal-02469747v1","doi":null,"is_oa":true,"landing_page_url":"https://cnam.hal.science/hal-02469747","pdf_url":null,"source":{"id":"https://openalex.org/S4306402512","display_name":"HAL","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I4405252976","host_organization_name":"Centre pour la Communication Scientifique Directe","host_organization_lineage":["https://openalex.org/I4405252976"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"33rd Conference on Neural Information Processing Systems (NeurIPS 2019), Dec 2019, Vancouver, Canada. pp.2898-2909","raw_type":"info:eu-repo/semantics/conferenceObject"},{"id":"doi:10.48550/arxiv.1910.04851","doi":"https://doi.org/10.48550/arxiv.1910.04851","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.1910.04851","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:1910.04851","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/1910.04851","pdf_url":"https://arxiv.org/pdf/1910.04851","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W2970565181.pdf","grobid_xml":"https://content.openalex.org/works/W2970565181.grobid-xml"},"referenced_works_count":45,"referenced_works":["https://openalex.org/W179875071","https://openalex.org/W582134693","https://openalex.org/W639708223","https://openalex.org/W1614298861","https://openalex.org/W1922655562","https://openalex.org/W1932198206","https://openalex.org/W1945616565","https://openalex.org/W2101946573","https://openalex.org/W2103328396","https://openalex.org/W2122217421","https://openalex.org/W2144746247","https://openalex.org/W2160815625","https://openalex.org/W2163605009","https://openalex.org/W2164411961","https://openalex.org/W2171943915","https://openalex.org/W2202310977","https://openalex.org/W2335728318","https://openalex.org/W2343285185","https://openalex.org/W2462906003","https://openalex.org/W2531327146","https://openalex.org/W2560321925","https://openalex.org/W2600383743","https://openalex.org/W2626967530","https://openalex.org/W2768927604","https://openalex.org/W2786712888","https://openalex.org/W2801681995","https://openalex.org/W2806075129","https://openalex.org/W2883712957","https://openalex.org/W2891303672","https://openalex.org/W2902727564","https://openalex.org/W2908048420","https://openalex.org/W2950170869","https://openalex.org/W2963207607","https://openalex.org/W2963351448","https://openalex.org/W2963505832","https://openalex.org/W2963613748","https://openalex.org/W2963677766","https://openalex.org/W2963693742","https://openalex.org/W2964153729","https://openalex.org/W2982378089","https://openalex.org/W3040318838","https://openalex.org/W3106250896","https://openalex.org/W3118608800","https://openalex.org/W4300124531","https://openalex.org/W4300126339"],"related_works":["https://openalex.org/W3093773656","https://openalex.org/W4287554138","https://openalex.org/W4389692358","https://openalex.org/W4312808544","https://openalex.org/W7151184484","https://openalex.org/W7214895126","https://openalex.org/W7203933572","https://openalex.org/W4415798588","https://openalex.org/W3035059588","https://openalex.org/W4395669171"],"abstract_inverted_index":{"Assessing":[0],"reliably":[1],"the":[2,18,38,47,55,70,76,93,111,114],"confidence":[3],"of":[4,14,21,72,113],"a":[5,29,97],"deep":[6],"neural":[7],"network":[8,120],"and":[9,123,130],"predicting":[10],"its":[11],"failures":[12],"is":[13,49,79],"primary":[15],"importance":[16],"for":[17,33,67,109,127,155],"practical":[19],"deployment":[20],"these":[22],"models.":[23],"In":[24],"this":[25,103],"paper,":[26],"we":[27,86],"propose":[28,87],"new":[30],"target":[31],"criterion":[32,91],"model":[34],"confidence,":[35],"corresponding":[36],"to":[37,88,102,145],"True":[39],"Class":[40,58],"Probability":[41,59],"(TCP).":[42],"We":[43,61,117,133],"show":[44,134],"how":[45],"using":[46],"TCP":[48,68,90],"more":[50],"suited":[51],"than":[52],"relying":[53],"on":[54,92],"classic":[56],"Maximum":[57],"(MCP).":[60],"provide":[62],"in":[63,69],"addition":[64],"theoretical":[65],"guarantees":[66],"context":[71],"failure":[73,156],"prediction.":[74,157],"Since":[75],"true":[77],"class":[78],"by":[80],"essence":[81],"unknown":[82],"at":[83],"test":[84],"time,":[85],"learn":[89],"training":[94],"set,":[95],"introducing":[96],"specific":[98],"learning":[99],"scheme":[100],"adapted":[101],"context.":[104],"Extensive":[105],"experiments":[106],"are":[107],"conducted":[108],"validating":[110],"relevance":[112],"proposed":[115],"approach.":[116],"study":[118],"various":[119],"architectures,":[121],"small":[122],"large":[124],"scale":[125],"datasets":[126],"image":[128],"classification":[129],"semantic":[131],"segmentation.":[132],"that":[135],"our":[136],"approach":[137],"consistently":[138],"outperforms":[139],"several":[140],"strong":[141],"methods,":[142],"from":[143],"MCP":[144],"Bayesian":[146],"uncertainty,":[147],"as":[148,150],"well":[149],"recent":[151],"approaches":[152],"specifically":[153],"designed":[154]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":24},{"year":2024,"cited_by_count":33},{"year":2023,"cited_by_count":46},{"year":2022,"cited_by_count":43},{"year":2021,"cited_by_count":42},{"year":2020,"cited_by_count":18},{"year":2019,"cited_by_count":1},{"year":2012,"cited_by_count":1}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2019-09-05T00:00:00"}