{"id":"https://openalex.org/W4406460276","doi":"https://doi.org/10.1109/bigdata62323.2024.10825398","title":"DeltaMask: Minimizing Communication in Federated Fine-Tuning of Vision Foundation Models","display_name":"DeltaMask: Minimizing Communication in Federated Fine-Tuning of Vision Foundation Models","publication_year":2024,"publication_date":"2024-12-15","ids":{"openalex":"https://openalex.org/W4406460276","doi":"https://doi.org/10.1109/bigdata62323.2024.10825398"},"language":"en","primary_location":{"id":"doi:10.1109/bigdata62323.2024.10825398","is_oa":false,"landing_page_url":"https://doi.org/10.1109/bigdata62323.2024.10825398","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE International Conference on Big Data (BigData)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://pure.tue.nl/ws/files/349909071/2024337517.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5051394672","display_name":"Vasileios Tsouvalas","orcid":"https://orcid.org/0000-0001-9143-1188"},"institutions":[{"id":"https://openalex.org/I83019370","display_name":"Eindhoven University of Technology","ror":"https://ror.org/02c2kyt77","country_code":"NL","type":"education","lineage":["https://openalex.org/I83019370"]}],"countries":["NL"],"is_corresponding":true,"raw_author_name":"Vasileios Tsouvalas","raw_affiliation_strings":["Eindhoven University of Technology,The Netherlands"],"affiliations":[{"raw_affiliation_string":"Eindhoven University of Technology,The Netherlands","institution_ids":["https://openalex.org/I83019370"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102985946","display_name":"Yuki M. Asano","orcid":"https://orcid.org/0000-0002-8533-4020"},"institutions":[{"id":"https://openalex.org/I3130238102","display_name":"Lutheran University of Applied Sciences Nuremberg","ror":"https://ror.org/01tbgce33","country_code":"DE","type":"education","lineage":["https://openalex.org/I3130238102"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Yuki M. Asano","raw_affiliation_strings":["University of Technology,Nuremberg,Germany"],"affiliations":[{"raw_affiliation_string":"University of Technology,Nuremberg,Germany","institution_ids":["https://openalex.org/I3130238102"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5011960578","display_name":"Aaqib Saeed","orcid":"https://orcid.org/0000-0003-1473-0322"},"institutions":[{"id":"https://openalex.org/I83019370","display_name":"Eindhoven University of Technology","ror":"https://ror.org/02c2kyt77","country_code":"NL","type":"education","lineage":["https://openalex.org/I83019370"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Aaqib Saeed","raw_affiliation_strings":["Eindhoven University of Technology,The Netherlands"],"affiliations":[{"raw_affiliation_string":"Eindhoven University of Technology,The Netherlands","institution_ids":["https://openalex.org/I83019370"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5051394672"],"corresponding_institution_ids":["https://openalex.org/I83019370"],"apc_list":null,"apc_paid":null,"fwci":0.2582,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.5851897,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":95},"biblio":{"volume":null,"issue":null,"first_page":"8042","last_page":"8051"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9983999729156494,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9983999729156494,"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/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.9965999722480774,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9962999820709229,"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/computer-science","display_name":"Computer science","score":0.7505934238433838},{"id":"https://openalex.org/keywords/foundation","display_name":"Foundation (evidence)","score":0.7294230461120605},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3564554452896118},{"id":"https://openalex.org/keywords/human\u2013computer-interaction","display_name":"Human\u2013computer interaction","score":0.3421766757965088}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7505934238433838},{"id":"https://openalex.org/C2780966255","wikidata":"https://www.wikidata.org/wiki/Q5474306","display_name":"Foundation (evidence)","level":2,"score":0.7294230461120605},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3564554452896118},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.3421766757965088},{"id":"https://openalex.org/C95457728","wikidata":"https://www.wikidata.org/wiki/Q309","display_name":"History","level":0,"score":0.0},{"id":"https://openalex.org/C166957645","wikidata":"https://www.wikidata.org/wiki/Q23498","display_name":"Archaeology","level":1,"score":0.0}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1109/bigdata62323.2024.10825398","is_oa":false,"landing_page_url":"https://doi.org/10.1109/bigdata62323.2024.10825398","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE International Conference on Big Data (BigData)","raw_type":"proceedings-article"},{"id":"pmh:oai:pure.tue.nl:openaire_cris_publications/0cd7b5a6-cd9e-4a15-9e8a-cf6df5f92f55","is_oa":false,"landing_page_url":"https://research.tue.nl/en/publications/0cd7b5a6-cd9e-4a15-9e8a-cf6df5f92f55","pdf_url":null,"source":{"id":"https://openalex.org/S4406922641","display_name":"TU/e Research Portal","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":"repository"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Tsouvalas, V, Asano, Y M & Saeed, A 2025, DeltaMask : Minimizing Communication in Federated Fine-Tuning of Vision Foundation Models. in W Ding, C-T Lu, F Wang, L Di, K Wu, J Huan, R Nambiar, J Li, F Ilievski, R Baeza-Yates & X Hu (eds), 2024 IEEE International Conference on Big Data, IEEE BigData 2024., 10825398, Institute of Electrical and Electronics Engineers, pp. 8042-8051, 2024 IEEE International Conference on Big Data 2024, IEEE BigData 2024, Washington DC, United States, 15/12/24. https://doi.org/10.1109/BigData62323.2024.10825398","raw_type":"info:eu-repo/semantics/publishedVersion"},{"id":"pmh:oai:pure.tue.nl:publications/0cd7b5a6-cd9e-4a15-9e8a-cf6df5f92f55","is_oa":true,"landing_page_url":"https://research.tue.nl/files/349909071/2024337517.pdf","pdf_url":"https://pure.tue.nl/ws/files/349909071/2024337517.pdf","source":{"id":"https://openalex.org/S4406922641","display_name":"TU/e Research Portal","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":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Tsouvalas, V, Asano, Y M & Saeed, A 2025, DeltaMask : Minimizing Communication in Federated Fine-Tuning of Vision Foundation Models. in W Ding, C-T Lu, F Wang, L Di, K Wu, J Huan, R Nambiar, J Li, F Ilievski, R Baeza-Yates & X Hu (eds), 2024 IEEE International Conference on Big Data, IEEE BigData 2024., 10825398, Institute of Electrical and Electronics Engineers, pp. 8042-8051, 2024 IEEE International Conference on Big Data 2024, IEEE BigData 2024, Washington DC, United States, 15/12/24. https://doi.org/10.1109/BigData62323.2024.10825398","raw_type":"info:eu-repo/semantics/publishedVersion"}],"best_oa_location":{"id":"pmh:oai:pure.tue.nl:publications/0cd7b5a6-cd9e-4a15-9e8a-cf6df5f92f55","is_oa":true,"landing_page_url":"https://research.tue.nl/files/349909071/2024337517.pdf","pdf_url":"https://pure.tue.nl/ws/files/349909071/2024337517.pdf","source":{"id":"https://openalex.org/S4406922641","display_name":"TU/e Research Portal","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":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Tsouvalas, V, Asano, Y M & Saeed, A 2025, DeltaMask : Minimizing Communication in Federated Fine-Tuning of Vision Foundation Models. in W Ding, C-T Lu, F Wang, L Di, K Wu, J Huan, R Nambiar, J Li, F Ilievski, R Baeza-Yates & X Hu (eds), 2024 IEEE International Conference on Big Data, IEEE BigData 2024., 10825398, Institute of Electrical and Electronics Engineers, pp. 8042-8051, 2024 IEEE International Conference on Big Data 2024, IEEE BigData 2024, Washington DC, United States, 15/12/24. https://doi.org/10.1109/BigData62323.2024.10825398","raw_type":"info:eu-repo/semantics/publishedVersion"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4406460276.pdf","grobid_xml":"https://content.openalex.org/works/W4406460276.grobid-xml"},"referenced_works_count":74,"referenced_works":["https://openalex.org/W12634471","https://openalex.org/W2138011018","https://openalex.org/W2335728318","https://openalex.org/W2535838896","https://openalex.org/W2590796488","https://openalex.org/W2606891064","https://openalex.org/W2750384547","https://openalex.org/W2769644379","https://openalex.org/W2774000609","https://openalex.org/W2791091755","https://openalex.org/W2805003733","https://openalex.org/W2892974366","https://openalex.org/W2911940095","https://openalex.org/W2930957955","https://openalex.org/W2963625110","https://openalex.org/W2970277060","https://openalex.org/W2990138404","https://openalex.org/W2994941531","https://openalex.org/W3011810846","https://openalex.org/W3034234149","https://openalex.org/W3048121440","https://openalex.org/W3095276777","https://openalex.org/W3101036738","https://openalex.org/W3118608800","https://openalex.org/W3126708604","https://openalex.org/W3169314795","https://openalex.org/W3214721897","https://openalex.org/W4214926007","https://openalex.org/W4285661751","https://openalex.org/W4286907821","https://openalex.org/W4287118122","https://openalex.org/W4287332481","https://openalex.org/W4297060071","https://openalex.org/W4299283926","https://openalex.org/W4300176521","https://openalex.org/W4300980304","https://openalex.org/W4320722432","https://openalex.org/W4366208220","https://openalex.org/W4381826961","https://openalex.org/W4382492045","https://openalex.org/W4385443277","https://openalex.org/W4401943506","https://openalex.org/W6683928519","https://openalex.org/W6702112450","https://openalex.org/W6703116779","https://openalex.org/W6728757088","https://openalex.org/W6733793881","https://openalex.org/W6744073472","https://openalex.org/W6746200960","https://openalex.org/W6746839373","https://openalex.org/W6751979845","https://openalex.org/W6755142261","https://openalex.org/W6761601108","https://openalex.org/W6773976177","https://openalex.org/W6776775496","https://openalex.org/W6781175393","https://openalex.org/W6781891088","https://openalex.org/W6784343361","https://openalex.org/W6787972765","https://openalex.org/W6788755940","https://openalex.org/W6790230083","https://openalex.org/W6790513813","https://openalex.org/W6791353385","https://openalex.org/W6794811910","https://openalex.org/W6811014117","https://openalex.org/W6839053766","https://openalex.org/W6840017308","https://openalex.org/W6840478408","https://openalex.org/W6845709166","https://openalex.org/W6849928207","https://openalex.org/W6851800889","https://openalex.org/W6853612481","https://openalex.org/W6854302354","https://openalex.org/W6858966089"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2381393187","https://openalex.org/W2332779545","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W2358060160","https://openalex.org/W2035483685"],"abstract_inverted_index":{"Foundation":[0],"Models":[1],"(FMs)":[2],"have":[3],"revolutionized":[4],"machine":[5],"learning":[6],"with":[7,58],"their":[8,16,30,59],"adaptability":[9],"and":[10,110,113,141,165],"high":[11],"performance":[12],"across":[13,138],"tasks;":[14],"yet,":[15],"integration":[17],"into":[18,121],"Federated":[19],"Learning":[20],"(FL)":[21],"is":[22],"challenging":[23],"due":[24],"to":[25,43,52,68,103,118],"substantial":[26],"communication":[27,69,154],"overhead":[28],"from":[29,130],"extensive":[31],"parameterization.":[32],"Current":[33],"communication-efficient":[34],"FL":[35,90],"strategies,":[36],"such":[37],"as":[38,148,150,160],"gradient":[39],"compression,":[40],"reduce":[41],"bitrates":[42,147],"around":[44],"1":[45,97],"bit-per-parameter":[46],"(bpp).":[47],"However,":[48],"these":[49,73],"approaches":[50],"fail":[51],"harness":[53],"the":[54],"characteristics":[55],"of":[56,62,169],"FMs,":[57],"large":[60],"number":[61],"parameters":[63],"still":[64],"posing":[65],"a":[66,82,122],"challenge":[67],"efficiency,":[70],"even":[71],"at":[72,91],"bitrate":[74],"regimes.":[75],"In":[76],"this":[77],"work,":[78],"we":[79],"present":[80],"DeltaMask,":[81],"novel":[83],"method":[84],"that":[85],"efficiently":[86,145],"fine-tunes":[87],"FMs":[88,109,158],"in":[89,115],"an":[92],"ultra-low":[93],"bitrate,":[94],"well":[95],"below":[96],"bpp.":[98],"DeltaMask":[99,144],"employs":[100],"stochastic":[101],"masking":[102],"detect":[104],"highly":[105],"effective":[106],"subnetworks":[107],"within":[108],"leverage":[111],"stochasticity":[112],"sparsity":[114],"client":[116],"masks":[117],"compress":[119],"updates":[120],"compact":[123],"grayscale":[124],"image":[125],"using":[126],"probabilistic":[127],"filters,":[128],"deviating":[129],"traditional":[131],"weight":[132],"training":[133],"approaches.":[134],"Our":[135],"comprehensive":[136],"evaluations":[137],"various":[139,170],"datasets":[140,164],"architectures":[142],"demonstrate":[143],"achieves":[146],"low":[149],"0.09":[151],"bpp,":[152],"enhancing":[153],"efficiency":[155],"while":[156],"maintaining":[157],"performance,":[159],"measured":[161],"on":[162],"8":[163],"5":[166],"pre-trained":[167],"models":[168],"network":[171],"architectures.":[172]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-03-12T08:34:05.389933","created_date":"2025-10-10T00:00:00"}