{"id":"https://openalex.org/W4401693251","doi":"https://doi.org/10.1109/isit57864.2024.10619279","title":"Hierarchical Generalization Bounds for Deep Neural Networks","display_name":"Hierarchical Generalization Bounds for Deep Neural Networks","publication_year":2024,"publication_date":"2024-07-07","ids":{"openalex":"https://openalex.org/W4401693251","doi":"https://doi.org/10.1109/isit57864.2024.10619279"},"language":"en","primary_location":{"id":"doi:10.1109/isit57864.2024.10619279","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/isit57864.2024.10619279","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 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/A5073481627","display_name":"Haiyun He","orcid":"https://orcid.org/0000-0002-1797-6101"},"institutions":[{"id":"https://openalex.org/I205783295","display_name":"Cornell University","ror":"https://ror.org/05bnh6r87","country_code":"US","type":"education","lineage":["https://openalex.org/I205783295"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Haiyun He","raw_affiliation_strings":["Cornell University,Center for Applied Mathematics,Ithaca,NY,USA"],"affiliations":[{"raw_affiliation_string":"Cornell University,Center for Applied Mathematics,Ithaca,NY,USA","institution_ids":["https://openalex.org/I205783295"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5032294826","display_name":"Christina Lee Yu","orcid":"https://orcid.org/0000-0002-2165-5220"},"institutions":[{"id":"https://openalex.org/I205783295","display_name":"Cornell University","ror":"https://ror.org/05bnh6r87","country_code":"US","type":"education","lineage":["https://openalex.org/I205783295"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Christina Lee Yu","raw_affiliation_strings":["School of ORIE, Cornell University,Ithaca,NY,USA"],"affiliations":[{"raw_affiliation_string":"School of ORIE, Cornell University,Ithaca,NY,USA","institution_ids":["https://openalex.org/I205783295"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5071112095","display_name":"Ziv Goldfeld","orcid":"https://orcid.org/0000-0003-3406-3950"},"institutions":[{"id":"https://openalex.org/I205783295","display_name":"Cornell University","ror":"https://ror.org/05bnh6r87","country_code":"US","type":"education","lineage":["https://openalex.org/I205783295"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ziv Goldfeld","raw_affiliation_strings":["School of ECE, Cornell University,Ithaca,NY,USA"],"affiliations":[{"raw_affiliation_string":"School of ECE, Cornell University,Ithaca,NY,USA","institution_ids":["https://openalex.org/I205783295"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5073481627"],"corresponding_institution_ids":["https://openalex.org/I205783295"],"apc_list":null,"apc_paid":null,"fwci":0.3626,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.65696166,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":94},"biblio":{"volume":null,"issue":null,"first_page":"2688","last_page":"2693"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.9162999987602234,"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/T10320","display_name":"Neural Networks and Applications","score":0.9162999987602234,"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":0.7759228348731995},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6766641139984131},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.573010265827179},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5545312166213989},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.5005631446838379},{"id":"https://openalex.org/keywords/theoretical-computer-science","display_name":"Theoretical computer science","score":0.33414924144744873},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.20952892303466797}],"concepts":[{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.7759228348731995},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6766641139984131},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.573010265827179},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5545312166213989},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.5005631446838379},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.33414924144744873},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.20952892303466797},{"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/isit57864.2024.10619279","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/isit57864.2024.10619279","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE International Symposium on Information Theory (ISIT)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1108386053","display_name":null,"funder_award_id":"CNS-1955997,CCF-2046018,DMS-2210368,CCF-2308446","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G3015205638","display_name":null,"funder_award_id":"FA9550-23-1-0301","funder_id":"https://openalex.org/F4320338279","funder_display_name":"Air Force Office of Scientific Research"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320338279","display_name":"Air Force Office of Scientific Research","ror":"https://ror.org/011e9bt93"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":26,"referenced_works":["https://openalex.org/W1585160083","https://openalex.org/W2014384147","https://openalex.org/W2566079294","https://openalex.org/W2732724430","https://openalex.org/W2791379569","https://openalex.org/W2887344700","https://openalex.org/W2912811302","https://openalex.org/W2951714068","https://openalex.org/W2963038205","https://openalex.org/W2963799213","https://openalex.org/W2963862692","https://openalex.org/W2976533594","https://openalex.org/W2979452771","https://openalex.org/W3019669224","https://openalex.org/W3081982129","https://openalex.org/W3100231902","https://openalex.org/W3106002857","https://openalex.org/W3170524081","https://openalex.org/W3198746530","https://openalex.org/W3203779267","https://openalex.org/W4225693694","https://openalex.org/W4287332259","https://openalex.org/W4293774175","https://openalex.org/W4299785371","https://openalex.org/W4310348270","https://openalex.org/W4394007220"],"related_works":["https://openalex.org/W3162204513","https://openalex.org/W2371138613","https://openalex.org/W2048963458","https://openalex.org/W43109613","https://openalex.org/W2359952343","https://openalex.org/W2239445980","https://openalex.org/W2080152487","https://openalex.org/W3083152911","https://openalex.org/W3022347918","https://openalex.org/W4200527723"],"abstract_inverted_index":{"Deep":[0],"neural":[1],"networks":[2],"(DNNs)":[3],"exhibit":[4],"an":[5],"exceptional":[6],"generalization":[7,31,41,59,95,100,144],"capability":[8],"in":[9,61,145],"practice.":[10],"This":[11],"work":[12],"aims":[13],"to":[14,107],"capture":[15],"the":[16,27,40,47,64,71,82,86,99,108,127,131],"effect":[17],"of":[18,29,63,88,110],"depth":[19],"and":[20,114,130],"its":[21],"potential":[22],"benefit":[23],"for":[24,126,142],"learning":[25],"within":[26,50],"paradigm":[28],"information-theoretic":[30],"bounds.":[32],"We":[33,102],"derive":[34],"two":[35],"novel":[36],"hierarchical":[37],"bounds":[38,106],"on":[39,46,81,119],"error":[42],"that":[43,91],"explicitly":[44],"depend":[45],"internal":[48],"representations":[49],"each":[51],"layer.":[52],"The":[53,75],"first":[54],"result,":[55],"is":[56,79],"a":[57,89,94,139],"layer-dependent":[58],"bound":[60],"terms":[62],"Kullback-Leibler":[65],"(KL)":[66],"divergence,":[67],"which":[68,78,97],"shrinks":[69],"as":[70,93],"layer":[72,90],"index":[73],"increases.":[74],"second":[76],"bound,":[77],"based":[80],"Wasserstein":[83,132],"distance,":[84],"implies":[85],"existence":[87],"serves":[92],"funnel,":[96],"minimizes":[98],"bound.":[101],"then":[103],"specialize":[104],"our":[105],"case":[109],"binary":[111],"Gaussian":[112],"classification,":[113],"present":[115],"analytic":[116],"expressions":[117],"dependent":[118],"weight":[120],"matrices":[121],"rank":[122],"or":[123],"certain":[124],"norms,":[125],"KL":[128],"divergence":[129],"bounds,":[133],"respectively.":[134],"Our":[135],"results":[136],"may":[137],"provide":[138],"new":[140],"perspective":[141],"understanding":[143],"deep":[146],"models.":[147]},"counts_by_year":[{"year":2024,"cited_by_count":1}],"updated_date":"2025-12-22T23:10:17.713674","created_date":"2025-10-10T00:00:00"}