{"id":"https://openalex.org/W4289655268","doi":"https://doi.org/10.1109/isit50566.2022.9834400","title":"Understanding Generalization via Leave-One-Out Conditional Mutual Information","display_name":"Understanding Generalization via Leave-One-Out Conditional Mutual Information","publication_year":2022,"publication_date":"2022-06-26","ids":{"openalex":"https://openalex.org/W4289655268","doi":"https://doi.org/10.1109/isit50566.2022.9834400"},"language":"en","primary_location":{"id":"doi:10.1109/isit50566.2022.9834400","is_oa":false,"landing_page_url":"https://doi.org/10.1109/isit50566.2022.9834400","pdf_url":null,"source":{"id":"https://openalex.org/S4363604560","display_name":"2022 IEEE International Symposium on Information Theory (ISIT)","issn_l":null,"issn":null,"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":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2022 IEEE International Symposium on Information Theory (ISIT)","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/A5061900418","display_name":"Mahdi Haghifam","orcid":"https://orcid.org/0000-0002-2719-9158"},"institutions":[{"id":"https://openalex.org/I4210127509","display_name":"Vector Institute","ror":"https://ror.org/03kqdja62","country_code":"CA","type":"facility","lineage":["https://openalex.org/I4210127509"]}],"countries":["CA"],"is_corresponding":true,"raw_author_name":"Mahdi Haghifam","raw_affiliation_strings":["Univ. Toronto,Electrical and Computer Eng","Electrical and Computer Eng., Univ. Toronto","Vector Institute"],"affiliations":[{"raw_affiliation_string":"Univ. Toronto,Electrical and Computer Eng","institution_ids":[]},{"raw_affiliation_string":"Electrical and Computer Eng., Univ. Toronto","institution_ids":[]},{"raw_affiliation_string":"Vector Institute","institution_ids":["https://openalex.org/I4210127509"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5047793114","display_name":"Shay Moran","orcid":"https://orcid.org/0000-0002-8662-2737"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]},{"id":"https://openalex.org/I174306211","display_name":"Technion \u2013 Israel Institute of Technology","ror":"https://ror.org/03qryx823","country_code":"IL","type":"education","lineage":["https://openalex.org/I174306211"]}],"countries":["IL","US"],"is_corresponding":false,"raw_author_name":"Shay Moran","raw_affiliation_strings":["Google Research","Technion"],"affiliations":[{"raw_affiliation_string":"Google Research","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"Technion","institution_ids":["https://openalex.org/I174306211"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5110275739","display_name":"Daniel M. Roy","orcid":null},"institutions":[{"id":"https://openalex.org/I4210127509","display_name":"Vector Institute","ror":"https://ror.org/03kqdja62","country_code":"CA","type":"facility","lineage":["https://openalex.org/I4210127509"]},{"id":"https://openalex.org/I4210157790","display_name":"Statistical Research (United States)","ror":"https://ror.org/058eg8775","country_code":"US","type":"company","lineage":["https://openalex.org/I4210157790"]}],"countries":["CA","US"],"is_corresponding":false,"raw_author_name":"Daniel M. Roy","raw_affiliation_strings":["Univ. Toronto,Electrical and Computer Eng","Statistical Sciences, Univ. Toronto","Electrical and Computer Eng., Univ. Toronto","Vector Institute"],"affiliations":[{"raw_affiliation_string":"Univ. Toronto,Electrical and Computer Eng","institution_ids":[]},{"raw_affiliation_string":"Statistical Sciences, Univ. Toronto","institution_ids":["https://openalex.org/I4210157790"]},{"raw_affiliation_string":"Electrical and Computer Eng., Univ. Toronto","institution_ids":[]},{"raw_affiliation_string":"Vector Institute","institution_ids":["https://openalex.org/I4210127509"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5056580090","display_name":"Gintare Karolina Dziugiate","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Gintare Karolina Dziugiate","raw_affiliation_strings":["Google Research"],"affiliations":[{"raw_affiliation_string":"Google Research","institution_ids":["https://openalex.org/I1291425158"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5061900418"],"corresponding_institution_ids":["https://openalex.org/I4210127509"],"apc_list":null,"apc_paid":null,"fwci":0.6233,"has_fulltext":false,"cited_by_count":6,"citation_normalized_percentile":{"value":0.67167636,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"2487","last_page":"2492"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12072","display_name":"Machine Learning and Algorithms","score":0.9980999827384949,"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/T12072","display_name":"Machine Learning and Algorithms","score":0.9980999827384949,"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/T13702","display_name":"Machine Learning in Healthcare","score":0.994700014591217,"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/T11303","display_name":"Bayesian Modeling and Causal Inference","score":0.9918000102043152,"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/generalization","display_name":"Generalization","score":0.7507544755935669},{"id":"https://openalex.org/keywords/bounded-function","display_name":"Bounded function","score":0.7260794043540955},{"id":"https://openalex.org/keywords/mutual-information","display_name":"Mutual information","score":0.6962254643440247},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5278539657592773},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.5183323621749878},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.46515852212905884},{"id":"https://openalex.org/keywords/population","display_name":"Population","score":0.42459186911582947},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.39106160402297974},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.3785894513130188},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3576950430870056},{"id":"https://openalex.org/keywords/theoretical-computer-science","display_name":"Theoretical computer science","score":0.30000606179237366}],"concepts":[{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.7507544755935669},{"id":"https://openalex.org/C34388435","wikidata":"https://www.wikidata.org/wiki/Q2267362","display_name":"Bounded function","level":2,"score":0.7260794043540955},{"id":"https://openalex.org/C152139883","wikidata":"https://www.wikidata.org/wiki/Q252973","display_name":"Mutual information","level":2,"score":0.6962254643440247},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5278539657592773},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.5183323621749878},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.46515852212905884},{"id":"https://openalex.org/C2908647359","wikidata":"https://www.wikidata.org/wiki/Q2625603","display_name":"Population","level":2,"score":0.42459186911582947},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.39106160402297974},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3785894513130188},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3576950430870056},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.30000606179237366},{"id":"https://openalex.org/C149923435","wikidata":"https://www.wikidata.org/wiki/Q37732","display_name":"Demography","level":1,"score":0.0},{"id":"https://openalex.org/C144024400","wikidata":"https://www.wikidata.org/wiki/Q21201","display_name":"Sociology","level":0,"score":0.0},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/isit50566.2022.9834400","is_oa":false,"landing_page_url":"https://doi.org/10.1109/isit50566.2022.9834400","pdf_url":null,"source":{"id":"https://openalex.org/S4363604560","display_name":"2022 IEEE International Symposium on Information Theory (ISIT)","issn_l":null,"issn":null,"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":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2022 IEEE International Symposium on Information Theory (ISIT)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/10","display_name":"Reduced inequalities","score":0.800000011920929}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":32,"referenced_works":["https://openalex.org/W1663583528","https://openalex.org/W2066688546","https://openalex.org/W2135912651","https://openalex.org/W2258658829","https://openalex.org/W2303608025","https://openalex.org/W2617539226","https://openalex.org/W2910589063","https://openalex.org/W2963007973","https://openalex.org/W2963103083","https://openalex.org/W3038014551","https://openalex.org/W3046866365","https://openalex.org/W3100231902","https://openalex.org/W3106002857","https://openalex.org/W3159649695","https://openalex.org/W3203779267","https://openalex.org/W4236362309","https://openalex.org/W4245577611","https://openalex.org/W4286858784","https://openalex.org/W4293583552","https://openalex.org/W4300167521","https://openalex.org/W6636906085","https://openalex.org/W6692563690","https://openalex.org/W6712926332","https://openalex.org/W6738074204","https://openalex.org/W6748601875","https://openalex.org/W6750582233","https://openalex.org/W6772932235","https://openalex.org/W6780309617","https://openalex.org/W6781432962","https://openalex.org/W6802511957","https://openalex.org/W6804060860","https://openalex.org/W7034108470"],"related_works":["https://openalex.org/W2466816617","https://openalex.org/W1970834875","https://openalex.org/W3174028392","https://openalex.org/W842936808","https://openalex.org/W3162204513","https://openalex.org/W2000517284","https://openalex.org/W2136503713","https://openalex.org/W2365318811","https://openalex.org/W2375330620","https://openalex.org/W2913710507"],"abstract_inverted_index":{"We":[0,36],"study":[1],"the":[2,6,27,42,60,93,97,108,117,126,134],"mutual":[3,44],"information":[4,45],"between":[5],"output":[7],"of":[8,21,41,47,64,96],"a":[9,19,70,100],"learning":[10,67,102,130],"algorithm":[11,49,68,103],"and":[12,51,76,111],"its":[13],"n":[14],"training":[15,28],"data,":[16],"conditional":[17,43],"on":[18],"supersample":[20],"n+1":[22],"i.i.d.":[23],"data":[24,29],"from":[25],"which":[26,54],"is":[30,116,122],"chosen":[31],"at":[32],"random":[33],"without":[34],"replacement.":[35],"show":[37],"that":[38,81,113,121],"this":[39],"variant":[40],"(CMI)":[46],"an":[48,89],"(Steinke":[50],"Zakynthinou,":[52],"2020),":[53],"we":[55,79,91],"dub":[56],"leave-one-out":[57,83],"CMI,":[58],"determines":[59],"mean":[61],"generalization":[62],"error":[63],"any":[65],"interpolating":[66,75],"with":[69],"bounded":[71,82],"loss":[72],"function.":[73],"For":[74],"non-interpolating":[77],"classifiers,":[78],"demonstrate":[80],"CMI":[84],"implies":[85],"generalization.":[86],"Finally,":[87],"as":[88],"application,":[90],"analyze":[92],"population":[94],"risk":[95],"One-Inclusion-Graph":[98],"Algorithm,":[99],"transductive":[101],"for":[104,129],"VC":[105,131],"classes":[106,132],"in":[107,133],"realizable":[109,135],"setting,":[110],"prove":[112],"our":[114],"framework":[115,120],"first":[118],"information-theoretic":[119],"able":[123],"to":[124],"achieve":[125],"optimal":[127],"bound":[128],"setting.":[136]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":3}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}