{"id":"https://openalex.org/W7138711875","doi":"https://doi.org/10.48550/arxiv.2603.14773","title":"HO-SFL: Hybrid-Order Split Federated Learning with Backprop-Free Clients and Dimension-Free Aggregation","display_name":"HO-SFL: Hybrid-Order Split Federated Learning with Backprop-Free Clients and Dimension-Free Aggregation","publication_year":2026,"publication_date":"2026-03-16","ids":{"openalex":"https://openalex.org/W7138711875","doi":"https://doi.org/10.48550/arxiv.2603.14773"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.14773","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.14773","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.14773","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5129652980","display_name":"Qiyuan Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Chen, Qiyuan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129654276","display_name":"Xian Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Xian","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129694574","display_name":"Yi Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Yi","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5129698051","display_name":"Xianhao Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Xianhao","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5129652980"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"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/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.4537000060081482,"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/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.4537000060081482,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.08070000261068344,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.06419999897480011,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/convergence","display_name":"Convergence (economics)","score":0.7365000247955322},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.6517000198364258},{"id":"https://openalex.org/keywords/modalities","display_name":"Modalities","score":0.5349000096321106},{"id":"https://openalex.org/keywords/enhanced-data-rates-for-gsm-evolution","display_name":"Enhanced Data Rates for GSM Evolution","score":0.4733000099658966},{"id":"https://openalex.org/keywords/federated-learning","display_name":"Federated learning","score":0.45989999175071716},{"id":"https://openalex.org/keywords/edge-device","display_name":"Edge device","score":0.4253000020980835},{"id":"https://openalex.org/keywords/backpropagation","display_name":"Backpropagation","score":0.3467999994754791}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8295000195503235},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.7365000247955322},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.6517000198364258},{"id":"https://openalex.org/C2779903281","wikidata":"https://www.wikidata.org/wiki/Q6888026","display_name":"Modalities","level":2,"score":0.5349000096321106},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.4733000099658966},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.4699000120162964},{"id":"https://openalex.org/C2992525071","wikidata":"https://www.wikidata.org/wiki/Q50818671","display_name":"Federated learning","level":2,"score":0.45989999175071716},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.43130001425743103},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.4253000020980835},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.35339999198913574},{"id":"https://openalex.org/C155032097","wikidata":"https://www.wikidata.org/wiki/Q798503","display_name":"Backpropagation","level":3,"score":0.3467999994754791},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.3433000147342682},{"id":"https://openalex.org/C93996380","wikidata":"https://www.wikidata.org/wiki/Q44127","display_name":"Server","level":2,"score":0.329800009727478},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.3224000036716461},{"id":"https://openalex.org/C57869625","wikidata":"https://www.wikidata.org/wiki/Q1783502","display_name":"Rate of convergence","level":3,"score":0.3203999996185303},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.31450000405311584},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.305400013923645},{"id":"https://openalex.org/C2779582901","wikidata":"https://www.wikidata.org/wiki/Q21013010","display_name":"Distributed learning","level":2,"score":0.2827000021934509}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.14773","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.14773","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.14773","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.14773","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Fine-tuning":[0],"large":[1],"models":[2],"on":[3,131],"edge":[4],"devices":[5],"is":[6],"severely":[7],"hindered":[8],"by":[9],"the":[10,55,65,88,114],"memory-intensive":[11],"backpropagation":[12],"(BP)":[13],"in":[14],"standard":[15],"frameworks":[16],"like":[17],"federated":[18],"learning":[19,57],"and":[20,135,153],"split":[21,56],"learning.":[22],"While":[23],"substituting":[24],"BP":[25,92],"with":[26],"zeroth-order":[27,80,119],"optimization":[28,66],"can":[29],"significantly":[30,149],"reduce":[31],"memory":[32,155],"footprints,":[33],"it":[34],"typically":[35],"suffers":[36],"from":[37],"prohibitively":[38],"degraded":[39],"convergence":[40,108,116,123,142],"speed.":[41],"To":[42],"resolve":[43],"this":[44],"dilemma,":[45],"we":[46,104],"propose":[47],"Hybrid-Order":[48],"Split":[49],"Federated":[50],"Learning":[51],"(HO-SFL).":[52],"By":[53],"reformulating":[54],"process":[58],"within":[59],"a":[60,106,122],"Lagrangian":[61],"framework,":[62],"HO-SFL":[63,112,140],"decouples":[64],"landscape:":[67],"The":[68],"server":[69],"performs":[70],"precise":[71],"first-order":[72,127,146],"updates":[73],"(i.e.,":[74],"BP),":[75],"whereas":[76],"clients":[77],"conduct":[78],"memory-efficient":[79],"optimization.":[81],"This":[82],"hybrid":[83],"design":[84],"not":[85],"only":[86],"eliminates":[87],"need":[89],"for":[90],"client-side":[91],"but":[93],"also":[94],"enables":[95],"dimension-free":[96],"model":[97],"aggregation,":[98],"drastically":[99],"lowering":[100],"communication":[101,151],"costs.":[102],"Crucially,":[103],"provide":[105],"theoretical":[107],"analysis,":[109],"demonstrating":[110],"that":[111,139],"mitigates":[113],"dimension-dependent":[115],"slowdown":[117],"of":[118],"optimization,":[120],"achieving":[121],"rate":[124],"comparable":[125,144],"to":[126,145],"methods.":[128],"Extensive":[129],"experiments":[130],"tasks":[132],"across":[133],"vision":[134],"language":[136],"modalities":[137],"validate":[138],"achieves":[141],"speeds":[143],"baselines":[147],"while":[148],"reducing":[150],"costs":[152],"client":[154],"footprints.":[156]},"counts_by_year":[],"updated_date":"2026-03-18T06:31:55.123368","created_date":"2026-03-18T00:00:00"}