{"id":"https://openalex.org/W4405906994","doi":"https://doi.org/10.1109/tai.2024.3523252","title":"Revisiting LARS for Large Batch Training Generalization of Neural Networks","display_name":"Revisiting LARS for Large Batch Training Generalization of Neural Networks","publication_year":2024,"publication_date":"2024-12-30","ids":{"openalex":"https://openalex.org/W4405906994","doi":"https://doi.org/10.1109/tai.2024.3523252"},"language":"en","primary_location":{"id":"doi:10.1109/tai.2024.3523252","is_oa":true,"landing_page_url":"https://doi.org/10.1109/tai.2024.3523252","pdf_url":null,"source":{"id":"https://openalex.org/S4210169448","display_name":"IEEE Transactions on Artificial Intelligence","issn_l":"2691-4581","issn":["2691-4581"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Artificial Intelligence","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://doi.org/10.1109/tai.2024.3523252","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5102646996","display_name":"Khoi Do","orcid":null},"institutions":[{"id":"https://openalex.org/I205274468","display_name":"Trinity College Dublin","ror":"https://ror.org/02tyrky19","country_code":"IE","type":"education","lineage":["https://openalex.org/I205274468"]}],"countries":["IE"],"is_corresponding":false,"raw_author_name":"Khoi Do","raw_affiliation_strings":["School of Computer Science and Statistics, Trinity College Dublin, Dublin 2, Ireland"],"raw_orcid":"https://orcid.org/0000-0002-2566-5637","affiliations":[{"raw_affiliation_string":"School of Computer Science and Statistics, Trinity College Dublin, Dublin 2, Ireland","institution_ids":["https://openalex.org/I205274468"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5054892339","display_name":"Minh-Duong Nguyen","orcid":"https://orcid.org/0000-0001-9856-2754"},"institutions":[{"id":"https://openalex.org/I4921948","display_name":"Pusan National University","ror":"https://ror.org/01an57a31","country_code":"KR","type":"education","lineage":["https://openalex.org/I4921948"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Minh-Duong Nguyen","raw_affiliation_strings":["Department of Information Convergence Engineering, Pusan National University, Busan, Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Information Convergence Engineering, Pusan National University, Busan, Republic of Korea","institution_ids":["https://openalex.org/I4921948"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5050879879","display_name":"Ti\u1ebfn Hoa Nguy\u1ec5n","orcid":"https://orcid.org/0000-0003-4743-5012"},"institutions":[{"id":"https://openalex.org/I94518387","display_name":"Hanoi University of Science and Technology","ror":"https://ror.org/04nyv3z04","country_code":"VN","type":"education","lineage":["https://openalex.org/I94518387"]}],"countries":["VN"],"is_corresponding":false,"raw_author_name":"Nguyen Tien Hoa","raw_affiliation_strings":["School of Electrical and Electronic Engineering, Hanoi University of Science and Technology, Hanoi, Vietnam"],"raw_orcid":"https://orcid.org/0000-0003-4743-5012","affiliations":[{"raw_affiliation_string":"School of Electrical and Electronic Engineering, Hanoi University of Science and Technology, Hanoi, Vietnam","institution_ids":["https://openalex.org/I94518387"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103245667","display_name":"Long Tran-Thanh","orcid":"https://orcid.org/0000-0003-1617-8316"},"institutions":[{"id":"https://openalex.org/I39555362","display_name":"University of Warwick","ror":"https://ror.org/01a77tt86","country_code":"GB","type":"education","lineage":["https://openalex.org/I39555362"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Long Tran-Thanh","raw_affiliation_strings":["School of Computer Science, University of Warwick, Coventry, U.K","School of Computer Science, University of Warwick, Coventry, United Kingdom"],"raw_orcid":"https://orcid.org/0000-0003-1617-8316","affiliations":[{"raw_affiliation_string":"School of Computer Science, University of Warwick, Coventry, U.K","institution_ids":["https://openalex.org/I39555362"]},{"raw_affiliation_string":"School of Computer Science, University of Warwick, Coventry, United Kingdom","institution_ids":["https://openalex.org/I39555362"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027558961","display_name":"Nguyen H. Tran","orcid":"https://orcid.org/0000-0001-7323-9213"},"institutions":[{"id":"https://openalex.org/I129604602","display_name":"The University of Sydney","ror":"https://ror.org/0384j8v12","country_code":"AU","type":"education","lineage":["https://openalex.org/I129604602"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Nguyen H. Tran","raw_affiliation_strings":["School of Computer Science, The University of Sydney, Darlington, NSW, Australia"],"raw_orcid":"https://orcid.org/0000-0001-7323-9213","affiliations":[{"raw_affiliation_string":"School of Computer Science, The University of Sydney, Darlington, NSW, Australia","institution_ids":["https://openalex.org/I129604602"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5062525719","display_name":"Quoc\u2010Viet Pham","orcid":"https://orcid.org/0000-0002-9485-9216"},"institutions":[{"id":"https://openalex.org/I205274468","display_name":"Trinity College Dublin","ror":"https://ror.org/02tyrky19","country_code":"IE","type":"education","lineage":["https://openalex.org/I205274468"]}],"countries":["IE"],"is_corresponding":false,"raw_author_name":"Quoc-Viet Pham","raw_affiliation_strings":["School of Computer Science and Statistics, Trinity College Dublin, Dublin 2, Ireland"],"raw_orcid":"https://orcid.org/0000-0002-9485-9216","affiliations":[{"raw_affiliation_string":"School of Computer Science and Statistics, Trinity College Dublin, Dublin 2, Ireland","institution_ids":["https://openalex.org/I205274468"]}]}],"institutions":[],"countries_distinct_count":5,"institutions_distinct_count":5,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.8335,"has_fulltext":true,"cited_by_count":3,"citation_normalized_percentile":{"value":0.78914786,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":97},"biblio":{"volume":"6","issue":"5","first_page":"1321","last_page":"1333"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.9822999835014343,"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":0.9822999835014343,"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":0.7748450040817261},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.6411562561988831},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.6257555484771729},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.49041101336479187},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.48925256729125977},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.24994361400604248},{"id":"https://openalex.org/keywords/geography","display_name":"Geography","score":0.10332146286964417}],"concepts":[{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.7748450040817261},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.6411562561988831},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.6257555484771729},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.49041101336479187},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.48925256729125977},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.24994361400604248},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.10332146286964417},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0},{"id":"https://openalex.org/C153294291","wikidata":"https://www.wikidata.org/wiki/Q25261","display_name":"Meteorology","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/tai.2024.3523252","is_oa":true,"landing_page_url":"https://doi.org/10.1109/tai.2024.3523252","pdf_url":null,"source":{"id":"https://openalex.org/S4210169448","display_name":"IEEE Transactions on Artificial Intelligence","issn_l":"2691-4581","issn":["2691-4581"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Artificial Intelligence","raw_type":"journal-article"},{"id":"pmh:oai:wrap.warwick.ac.uk:189140","is_oa":true,"landing_page_url":"https://doi.org/10.1109/TAI.2024.3523252","pdf_url":"https://wrap.warwick.ac.uk/id/eprint/189140/1/10817779.pdf","source":{"id":"https://openalex.org/S4306400665","display_name":"Warwick Research Archive Portal (University of Warwick)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I39555362","host_organization_name":"University of Warwick","host_organization_lineage":["https://openalex.org/I39555362"],"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":"Journal Article"}],"best_oa_location":{"id":"doi:10.1109/tai.2024.3523252","is_oa":true,"landing_page_url":"https://doi.org/10.1109/tai.2024.3523252","pdf_url":null,"source":{"id":"https://openalex.org/S4210169448","display_name":"IEEE Transactions on Artificial Intelligence","issn_l":"2691-4581","issn":["2691-4581"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Artificial Intelligence","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":48,"referenced_works":["https://openalex.org/W1677182931","https://openalex.org/W2009797711","https://openalex.org/W2108598243","https://openalex.org/W2134120396","https://openalex.org/W2155893237","https://openalex.org/W2194775991","https://openalex.org/W2402144811","https://openalex.org/W2912811302","https://openalex.org/W2963016543","https://openalex.org/W2963743626","https://openalex.org/W2970971581","https://openalex.org/W2982083293","https://openalex.org/W3048584318","https://openalex.org/W3094502228","https://openalex.org/W3096609285","https://openalex.org/W3174135168","https://openalex.org/W3175277764","https://openalex.org/W4205816953","https://openalex.org/W4232753567","https://openalex.org/W4239640448","https://openalex.org/W4384918448","https://openalex.org/W4390874575","https://openalex.org/W4401307635","https://openalex.org/W6631943919","https://openalex.org/W6635810480","https://openalex.org/W6726497184","https://openalex.org/W6726983090","https://openalex.org/W6736583452","https://openalex.org/W6738534199","https://openalex.org/W6739622702","https://openalex.org/W6744513255","https://openalex.org/W6746419290","https://openalex.org/W6753209298","https://openalex.org/W6756358366","https://openalex.org/W6761074043","https://openalex.org/W6762287338","https://openalex.org/W6762718338","https://openalex.org/W6765560323","https://openalex.org/W6769303498","https://openalex.org/W6779977557","https://openalex.org/W6781318954","https://openalex.org/W6785738291","https://openalex.org/W6786016958","https://openalex.org/W6787972765","https://openalex.org/W6791742336","https://openalex.org/W6795754764","https://openalex.org/W6798077982","https://openalex.org/W6846194447"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W230091440","https://openalex.org/W2390279801","https://openalex.org/W2233261550","https://openalex.org/W3162204513","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345"],"abstract_inverted_index":{"This":[0],"article":[1],"investigates":[2],"large":[3],"batch":[4],"training":[5,86],"techniques":[6],"using":[7],"layer-wise":[8],"adaptive":[9],"scaling":[10],"ratio":[11,42],"(LARS)":[12],"across":[13],"diverse":[14],"settings.":[15],"In":[16,131],"particular,":[17],"we":[18,67],"first":[19],"show":[20],"that":[21,76,113],"a":[22,45,73,80],"state-of-the-art":[23],"technique,":[24],"called":[25],"LARS":[26,71,104,117],"with":[27,79,123],"the":[28,50,88],"warm-up,":[29],"tends":[30],"to":[31,40,103,125,139],"be":[32],"trapped":[33],"in":[34,49,87,107,120,128],"sharp":[35,61,98],"minimizers":[36],"early":[37,95],"on":[38],"due":[39],"redundant":[41],"scaling.":[43],"Additionally,":[44],"fixed":[46],"steep":[47],"decline":[48],"latter":[51],"phase":[52],"restricts":[53],"deep":[54],"neural":[55],"networks":[56],"from":[57],"effectively":[58],"navigating":[59],"early-phase":[60],"minimizers.":[62],"To":[63],"address":[64],"these":[65],"issues,":[66],"propose":[68],"time":[69],"varying":[70],"(TVLARS),":[72],"novel":[74],"algorithm":[75],"replaces":[77],"warm-up":[78],"configurable":[81],"sigmoid-like":[82],"function":[83],"for":[84,105],"robust":[85],"initial":[89],"phase.":[90],"TVLARS":[91,114,136],"promotes":[92],"gradient":[93],"exploration":[94],"on,":[96],"surpassing":[97],"optimizers":[99],"and":[100,118],"gradually":[101],"transitioning":[102],"robustness":[106],"later":[108],"stages.":[109],"Extensive":[110],"experiments":[111],"demonstrate":[112],"consistently":[115],"outperforms":[116],"LAMB":[119],"most":[121],"cases,":[122,135],"up":[124,138],"2%":[126],"improvement":[127],"classification":[129],"scenarios.":[130],"all":[132],"self-supervised":[133],"learning":[134],"achieves":[137],"10%":[140],"performance":[141],"improvement.":[142],"Our":[143],"implementation":[144],"is":[145],"available":[146],"at":[147],"<uri":[148],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[149],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">https://github.com/KhoiDOO/tvlars</uri>.":[150]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":2}],"updated_date":"2025-12-21T23:12:01.093139","created_date":"2025-10-10T00:00:00"}