{"id":"https://openalex.org/W4405727029","doi":"https://doi.org/10.3233/faia241456","title":"Enhancing Transformer Training Efficiency with Dynamic Dropout","display_name":"Enhancing Transformer Training Efficiency with Dynamic Dropout","publication_year":2024,"publication_date":"2024-12-20","ids":{"openalex":"https://openalex.org/W4405727029","doi":"https://doi.org/10.3233/faia241456"},"language":"en","primary_location":{"id":"doi:10.3233/faia241456","is_oa":true,"landing_page_url":"https://doi.org/10.3233/faia241456","pdf_url":null,"source":{"id":"https://openalex.org/S4210201731","display_name":"Frontiers in artificial intelligence and applications","issn_l":"0922-6389","issn":["0922-6389","1879-8314"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":"cc-by-nc","license_id":"https://openalex.org/licenses/cc-by-nc","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Frontiers in Artificial Intelligence and Applications","raw_type":"book-chapter"},"type":"book-chapter","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://doi.org/10.3233/faia241456","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Hanrui Yan","orcid":null},"institutions":[{"id":"https://openalex.org/I4210104631","display_name":"Transcend Engineering (United States)","ror":"https://ror.org/028mbk641","country_code":"US","type":"company","lineage":["https://openalex.org/I4210104631"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Hanrui Yan","raw_affiliation_strings":["ASCENDING inc., Fairfax VA 22031, USA"],"affiliations":[{"raw_affiliation_string":"ASCENDING inc., Fairfax VA 22031, USA","institution_ids":["https://openalex.org/I4210104631"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5020397061","display_name":"Dan Shao","orcid":"https://orcid.org/0000-0002-7429-3569"},"institutions":[{"id":"https://openalex.org/I4210104631","display_name":"Transcend Engineering (United States)","ror":"https://ror.org/028mbk641","country_code":"US","type":"company","lineage":["https://openalex.org/I4210104631"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Dan Shao","raw_affiliation_strings":["ASCENDING inc., Fairfax VA 22031, USA"],"affiliations":[{"raw_affiliation_string":"ASCENDING inc., Fairfax VA 22031, USA","institution_ids":["https://openalex.org/I4210104631"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4210104631"],"apc_list":null,"apc_paid":null,"fwci":2.9351,"has_fulltext":true,"cited_by_count":7,"citation_normalized_percentile":{"value":0.95264198,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":94,"max":98},"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.9937000274658203,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.9937000274658203,"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":0.9907000064849854,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.9866999983787537,"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/dropout","display_name":"Dropout (neural networks)","score":0.5959771871566772},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5946455001831055},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5862143039703369},{"id":"https://openalex.org/keywords/regularization","display_name":"Regularization (linguistics)","score":0.5818328857421875},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.5532768964767456},{"id":"https://openalex.org/keywords/schedule","display_name":"Schedule","score":0.43461689352989197},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.2885814309120178},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.2736174464225769},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.2307824194431305},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.19704791903495789}],"concepts":[{"id":"https://openalex.org/C2776145597","wikidata":"https://www.wikidata.org/wiki/Q25339462","display_name":"Dropout (neural networks)","level":2,"score":0.5959771871566772},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5946455001831055},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5862143039703369},{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.5818328857421875},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5532768964767456},{"id":"https://openalex.org/C68387754","wikidata":"https://www.wikidata.org/wiki/Q7271585","display_name":"Schedule","level":2,"score":0.43461689352989197},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2885814309120178},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2736174464225769},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.2307824194431305},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.19704791903495789},{"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/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.3233/faia241456","is_oa":true,"landing_page_url":"https://doi.org/10.3233/faia241456","pdf_url":null,"source":{"id":"https://openalex.org/S4210201731","display_name":"Frontiers in artificial intelligence and applications","issn_l":"0922-6389","issn":["0922-6389","1879-8314"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":"cc-by-nc","license_id":"https://openalex.org/licenses/cc-by-nc","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Frontiers in Artificial Intelligence and Applications","raw_type":"book-chapter"}],"best_oa_location":{"id":"doi:10.3233/faia241456","is_oa":true,"landing_page_url":"https://doi.org/10.3233/faia241456","pdf_url":null,"source":{"id":"https://openalex.org/S4210201731","display_name":"Frontiers in artificial intelligence and applications","issn_l":"0922-6389","issn":["0922-6389","1879-8314"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":"cc-by-nc","license_id":"https://openalex.org/licenses/cc-by-nc","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Frontiers in Artificial Intelligence and Applications","raw_type":"book-chapter"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":5,"referenced_works":["https://openalex.org/W2970971581","https://openalex.org/W4387109367","https://openalex.org/W4388186135","https://openalex.org/W4400770778","https://openalex.org/W4401227496"],"related_works":["https://openalex.org/W3082178636","https://openalex.org/W2782041652","https://openalex.org/W2612657834","https://openalex.org/W2392157706","https://openalex.org/W2599192953","https://openalex.org/W1987310671","https://openalex.org/W4206178588","https://openalex.org/W3094491777","https://openalex.org/W3214715529","https://openalex.org/W4287635093"],"abstract_inverted_index":{"We":[0],"introduce":[1],"Dynamic":[2,92,125],"Dropout,":[3],"a":[4,61,103,107,128],"novel":[5],"regularization":[6,38],"technique":[7,130],"designed":[8],"to":[9,59,102],"enhance":[10],"the":[11,20,34,56,86,117,122],"training":[12,25,70,96,132],"efficiency":[13,100],"of":[14,36,124],"Transformer":[15,134],"models":[16],"by":[17],"dynamically":[18],"adjusting":[19],"dropout":[21,63,67,109],"rate":[22,64],"based":[23],"on":[24,85],"epochs":[26],"or":[27],"validation":[28,80,112],"loss":[29],"improvements.":[30],"This":[31],"approach":[32],"addresses":[33],"challenge":[35],"balancing":[37],"and":[39,49,65,79,97],"model":[40,58,105],"capacity,":[41],"which":[42],"is":[43],"crucial":[44],"for":[45,131],"achieving":[46],"fast":[47],"convergence":[48],"high":[50],"performance.":[51],"Our":[52],"method":[53],"involves":[54],"modifying":[55],"GPT":[57],"accept":[60],"variable":[62],"updating":[66],"layers":[68],"during":[69],"using":[71],"schedules":[72],"such":[73],"as":[74,127],"linear":[75],"decay,":[76,78],"exponential":[77],"loss-based":[81,113],"adjustments.":[82],"Extensive":[83],"experiments":[84],"Shakespeare":[87],"char":[88],"dataset":[89],"demonstrate":[90],"that":[91],"Dropout":[93,126],"significantly":[94],"accelerates":[95],"improves":[98],"inference":[99],"compared":[101],"baseline":[104],"with":[106],"fixed":[108],"rate.":[110],"The":[111],"adjustment":[114],"schedule":[115],"provided":[116],"best":[118],"overall":[119],"performance,":[120],"highlighting":[121],"potential":[123],"valuable":[129],"large-scale":[133],"models.":[135]},"counts_by_year":[{"year":2025,"cited_by_count":5},{"year":2024,"cited_by_count":2}],"updated_date":"2026-04-21T08:09:41.155169","created_date":"2025-10-10T00:00:00"}