{"id":"https://openalex.org/W3022893496","doi":"https://doi.org/10.1109/ciss48834.2020.1570627167","title":"Exploring the Role of Loss Functions in Multiclass Classification","display_name":"Exploring the Role of Loss Functions in Multiclass Classification","publication_year":2020,"publication_date":"2020-03-01","ids":{"openalex":"https://openalex.org/W3022893496","doi":"https://doi.org/10.1109/ciss48834.2020.1570627167","mag":"3022893496"},"language":"en","primary_location":{"id":"doi:10.1109/ciss48834.2020.1570627167","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ciss48834.2020.1570627167","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2020 54th Annual Conference on Information Sciences and Systems (CISS)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5071823499","display_name":"Ahmet Demirkaya","orcid":"https://orcid.org/0000-0002-5859-6375"},"institutions":[{"id":"https://openalex.org/I168864056","display_name":"Bilkent University","ror":"https://ror.org/02vh8a032","country_code":"TR","type":"education","lineage":["https://openalex.org/I168864056"]}],"countries":["TR"],"is_corresponding":true,"raw_author_name":"Ahmet Demirkaya","raw_affiliation_strings":["Dept. of Electrical & Electronics Engineering, Bilkent University, Istanbul, Turkey"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Dept. of Electrical & Electronics Engineering, Bilkent University, Istanbul, Turkey","institution_ids":["https://openalex.org/I168864056"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5061832197","display_name":"Jiasi Chen","orcid":"https://orcid.org/0000-0001-9923-9027"},"institutions":[{"id":"https://openalex.org/I103635307","display_name":"University of California, Riverside","ror":"https://ror.org/03nawhv43","country_code":"US","type":"education","lineage":["https://openalex.org/I103635307"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jiasi Chen","raw_affiliation_strings":["Dept. of Computer Science & Engineering, University of California, Riverside, Riverside, CA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Dept. of Computer Science & Engineering, University of California, Riverside, Riverside, CA, USA","institution_ids":["https://openalex.org/I103635307"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5050547472","display_name":"Samet Oymak","orcid":"https://orcid.org/0000-0001-5203-0752"},"institutions":[{"id":"https://openalex.org/I103635307","display_name":"University of California, Riverside","ror":"https://ror.org/03nawhv43","country_code":"US","type":"education","lineage":["https://openalex.org/I103635307"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Samet Oymak","raw_affiliation_strings":["Dept. of Electrical & Computer Engineering, University of California, Riverside, Riverside, CA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Dept. of Electrical & Computer Engineering, University of California, Riverside, Riverside, CA, USA","institution_ids":["https://openalex.org/I103635307"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5071823499"],"corresponding_institution_ids":["https://openalex.org/I168864056"],"apc_list":null,"apc_paid":null,"fwci":2.1575,"has_fulltext":false,"cited_by_count":39,"citation_normalized_percentile":{"value":0.89523368,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"5"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10057","display_name":"Face and Expression Recognition","score":0.9962000250816345,"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"}},"topics":[{"id":"https://openalex.org/T10057","display_name":"Face and Expression Recognition","score":0.9962000250816345,"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/T11652","display_name":"Imbalanced Data Classification Techniques","score":0.9868999719619751,"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/T10320","display_name":"Neural Networks and Applications","score":0.9854000210762024,"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/cross-entropy","display_name":"Cross entropy","score":0.685955822467804},{"id":"https://openalex.org/keywords/quadratic-equation","display_name":"Quadratic equation","score":0.6838551163673401},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6619486808776855},{"id":"https://openalex.org/keywords/parameterized-complexity","display_name":"Parameterized complexity","score":0.5970300436019897},{"id":"https://openalex.org/keywords/entropy","display_name":"Entropy (arrow of time)","score":0.4868128001689911},{"id":"https://openalex.org/keywords/boosting","display_name":"Boosting (machine learning)","score":0.47175461053848267},{"id":"https://openalex.org/keywords/heuristics","display_name":"Heuristics","score":0.45943209528923035},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.44021520018577576},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.4211835265159607},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.41756510734558105},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.3675438165664673},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.2887251675128937},{"id":"https://openalex.org/keywords/principle-of-maximum-entropy","display_name":"Principle of maximum entropy","score":0.20023813843727112}],"concepts":[{"id":"https://openalex.org/C167981619","wikidata":"https://www.wikidata.org/wiki/Q1685498","display_name":"Cross entropy","level":3,"score":0.685955822467804},{"id":"https://openalex.org/C129844170","wikidata":"https://www.wikidata.org/wiki/Q41299","display_name":"Quadratic equation","level":2,"score":0.6838551163673401},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6619486808776855},{"id":"https://openalex.org/C165464430","wikidata":"https://www.wikidata.org/wiki/Q1570441","display_name":"Parameterized complexity","level":2,"score":0.5970300436019897},{"id":"https://openalex.org/C106301342","wikidata":"https://www.wikidata.org/wiki/Q4117933","display_name":"Entropy (arrow of time)","level":2,"score":0.4868128001689911},{"id":"https://openalex.org/C46686674","wikidata":"https://www.wikidata.org/wiki/Q466303","display_name":"Boosting (machine learning)","level":2,"score":0.47175461053848267},{"id":"https://openalex.org/C127705205","wikidata":"https://www.wikidata.org/wiki/Q5748245","display_name":"Heuristics","level":2,"score":0.45943209528923035},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.44021520018577576},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4211835265159607},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.41756510734558105},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.3675438165664673},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2887251675128937},{"id":"https://openalex.org/C9679016","wikidata":"https://www.wikidata.org/wiki/Q1417473","display_name":"Principle of maximum entropy","level":2,"score":0.20023813843727112},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/ciss48834.2020.1570627167","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ciss48834.2020.1570627167","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2020 54th Annual Conference on Information Sciences and Systems (CISS)","raw_type":"proceedings-article"},{"id":"pmh:oai:repository.bilkent.edu.tr:11693/54981","is_oa":false,"landing_page_url":"http://hdl.handle.net/11693/54981","pdf_url":null,"source":{"id":"https://openalex.org/S4306400079","display_name":"Bilkent University Institutional Repository (Bilkent University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I168864056","host_organization_name":"Bilkent University","host_organization_lineage":["https://openalex.org/I168864056"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"54th Annual Conference on Information Sciences and Systems, CISS 2020","raw_type":"Conference Paper"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1302182869","display_name":"NeTS: Small: Support for Interactive AR/VR Video: Learning and Optimizing at the Network Edge","funder_award_id":"1817216","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G2603905188","display_name":null,"funder_award_id":"1932254","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G3309364101","display_name":null,"funder_award_id":"CNS-1817216","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":46,"referenced_works":["https://openalex.org/W1480538416","https://openalex.org/W1553004968","https://openalex.org/W1667165204","https://openalex.org/W1988115241","https://openalex.org/W2006626130","https://openalex.org/W2103496339","https://openalex.org/W2120725344","https://openalex.org/W2163605009","https://openalex.org/W2194775991","https://openalex.org/W2406273144","https://openalex.org/W2461743311","https://openalex.org/W2553576544","https://openalex.org/W2765329844","https://openalex.org/W2807842867","https://openalex.org/W2899790086","https://openalex.org/W2900959181","https://openalex.org/W2905800572","https://openalex.org/W2909326933","https://openalex.org/W2942228371","https://openalex.org/W2949960976","https://openalex.org/W2952126211","https://openalex.org/W2962698540","https://openalex.org/W2963038274","https://openalex.org/W2964137095","https://openalex.org/W2965497096","https://openalex.org/W2965772785","https://openalex.org/W2970423380","https://openalex.org/W2995625976","https://openalex.org/W3011054152","https://openalex.org/W3137695714","https://openalex.org/W4288279338","https://openalex.org/W6628642759","https://openalex.org/W6684191040","https://openalex.org/W6714078828","https://openalex.org/W6718527420","https://openalex.org/W6752127812","https://openalex.org/W6752453940","https://openalex.org/W6756001544","https://openalex.org/W6756137178","https://openalex.org/W6757685903","https://openalex.org/W6757889297","https://openalex.org/W6763485134","https://openalex.org/W6763845970","https://openalex.org/W6764891227","https://openalex.org/W6766222583","https://openalex.org/W6766887723"],"related_works":["https://openalex.org/W2051058708","https://openalex.org/W154868527","https://openalex.org/W1494268238","https://openalex.org/W1983207144","https://openalex.org/W2490706771","https://openalex.org/W2480116122","https://openalex.org/W1976468483","https://openalex.org/W1516574938","https://openalex.org/W2563912921","https://openalex.org/W3040748430"],"abstract_inverted_index":{"Cross-entropy":[0],"is":[1,30,97],"the":[2,47,63,88,109,145,153,158,164,171,199,218,238],"de-facto":[3],"loss":[4,22,37,66,81,135,185,194,222,234],"function":[5,38],"in":[6,68,87,113,137,242],"modern":[7],"classification":[8],"tasks":[9,70],"that":[10,79,93,217],"involve":[11],"distinguishing":[12],"hundreds":[13],"or":[14],"even":[15],"thousands":[16],"of":[17,65,75,90,95,111,201,229],"classes.":[18,76],"To":[19],"design":[20],"better":[21,50,102],"functions":[23,67,82],"for":[24,40,101],"new":[25],"machine":[26],"learning":[27,69],"tasks,":[28],"it":[29,122],"critical":[31],"to":[32,144,149,235],"understand":[33],"what":[34,45],"makes":[35,46],"a":[36,41,72,98,117,138,180,226,243],"suitable":[39],"problem.":[42],"For":[43,174],"instance,":[44],"cross":[48],"entropy":[49],"than":[51],"other":[52],"alternatives":[53],"such":[54],"as":[55,198],"quadratic":[56,146,193,221,233],"loss?":[57],"In":[58],"this":[59],"work,":[60],"we":[61,178],"discuss":[62],"role":[64],"with":[71,157,209,220,225],"large":[73,85],"number":[74,200,228],"We":[77,130,215],"hypothesize":[78],"different":[80],"can":[83,186],"have":[84],"variability":[86],"difficulty":[89],"optimization":[91],"and":[92,125,167,203],"simplicity":[94],"training":[96,165,190],"key":[99],"catalyst":[100],"test-time":[103],"performance.":[104,173,246],"Our":[105],"intuition":[106],"draws":[107],"from":[108],"success":[110],"over-parameterization":[112,161],"deep":[114],"learning:":[115],"As":[116],"model":[118],"has":[119,195],"more":[120,140],"parameters,":[121],"trains":[123],"faster":[124],"achieves":[126],"higher":[127],"test":[128,172],"accuracy.":[129],"argue":[131],"that,":[132],"effectively,":[133],"cross-entropy":[134,184],"results":[136,241],"much":[139],"over-parameterized":[141],"problem":[142],"compared":[143],"loss,":[147,191],"thanks":[148],"its":[150],"emphasis":[151],"on":[152,237],"correct":[154,239],"class":[155,204,240],"(associated":[156],"label).":[159],"Such":[160],"drastically":[162,244],"simplifies":[163],"process":[166],"ends":[168],"up":[169],"boosting":[170],"separable":[175],"mixture":[176],"models,":[177],"provide":[179],"separation":[181],"result":[182],"where":[183],"always":[187],"achieve":[188],"small":[189],"whereas":[192],"diminishing":[196],"benefit":[197],"classes":[202],"correlations":[205],"increase.":[206],"Numerical":[207],"experiments":[208],"CIFAR":[210],"100":[211],"corroborate":[212],"our":[213],"results.":[214],"show":[216],"accuracy":[219],"disproportionately":[223],"degrades":[224],"growing":[227],"classes;":[230],"however,":[231],"encouraging":[232],"focus":[236],"improved":[245]},"counts_by_year":[{"year":2025,"cited_by_count":10},{"year":2024,"cited_by_count":7},{"year":2023,"cited_by_count":3},{"year":2022,"cited_by_count":7},{"year":2021,"cited_by_count":10},{"year":2020,"cited_by_count":2}],"updated_date":"2026-05-05T08:41:31.759640","created_date":"2025-10-10T00:00:00"}