{"id":"https://openalex.org/W7127568340","doi":"https://doi.org/10.48550/arxiv.2602.02696","title":"NSC-SL: A Bandwidth-Aware Neural Subspace Compression for Communication-Efficient Split Learning","display_name":"NSC-SL: A Bandwidth-Aware Neural Subspace Compression for Communication-Efficient Split Learning","publication_year":2026,"publication_date":"2026-02-02","ids":{"openalex":"https://openalex.org/W7127568340","doi":"https://doi.org/10.48550/arxiv.2602.02696"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2602.02696","doi":"https://doi.org/10.48550/arxiv.2602.02696","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.02696","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"Preprint"},"volume":null,"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2602.02696","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5090538195","display_name":"Zhen Fang","orcid":"https://orcid.org/0000-0001-8482-8013","observed_orcids":["https://orcid.org/0000-0001-8482-8013"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fang, Zhen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125069005","display_name":"Miao Yang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Miao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5073247041","display_name":"Zehang Lin","orcid":"https://orcid.org/0009-0008-1601-4729","observed_orcids":["https://orcid.org/0009-0008-1601-4729"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lin, Zehang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5122022864","display_name":"Zheng Lin","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lin, Zheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135713437","display_name":"Zihan Fang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fang, Zihan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5123012476","display_name":"Zongyuan Zhang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Zongyuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5122994919","display_name":"Tianyang Duan","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Duan, Tianyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5050270789","display_name":"Dong Huang","orcid":"https://orcid.org/0000-0003-3923-8828","observed_orcids":["https://orcid.org/0000-0003-3923-8828"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Dong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5036478981","display_name":"Shunzhi Zhu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhu, Shunzhi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"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":0.9106000065803528,"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":0.9106000065803528,"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":0.06989999860525131,"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":0.002199999988079071,"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/split-learning","display_name":"split learning","score":0.8970000147819519},{"id":"https://openalex.org/keywords/neural-network-compression","display_name":"neural network compression","score":0.7860000133514404},{"id":"https://openalex.org/keywords/communication-efficiency","display_name":"communication efficiency","score":0.8130000233650208},{"id":"https://openalex.org/keywords/low-rank-approximation","display_name":"low-rank approximation","score":0.9570000171661377},{"id":"https://openalex.org/keywords/singular-value-distribution","display_name":"singular value distribution","score":0.8330000042915344},{"id":"https://openalex.org/keywords/adaptive-rank-selection","display_name":"adaptive rank selection","score":0.7379999756813049},{"id":"https://openalex.org/keywords/tensor-decomposition","display_name":"tensor decomposition","score":0.9110000133514404},{"id":"https://openalex.org/keywords/distributed-learning","display_name":"distributed learning","score":0.8360000252723694}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7667999863624573},{"id":"https://openalex.org/C22789450","wikidata":"https://www.wikidata.org/wiki/Q420904","display_name":"Singular value decomposition","level":2,"score":0.5315999984741211},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.49630001187324524},{"id":"https://openalex.org/C32834561","wikidata":"https://www.wikidata.org/wiki/Q660730","display_name":"Subspace topology","level":2,"score":0.48829999566078186},{"id":"https://openalex.org/C155512373","wikidata":"https://www.wikidata.org/wiki/Q287450","display_name":"Residual","level":2,"score":0.48750001192092896},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.4862000048160553},{"id":"https://openalex.org/C180016635","wikidata":"https://www.wikidata.org/wiki/Q2712821","display_name":"Compression (physics)","level":2,"score":0.48500001430511475},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.477400004863739},{"id":"https://openalex.org/C81081738","wikidata":"https://www.wikidata.org/wiki/Q55542","display_name":"Lossless compression","level":3,"score":0.47380000352859497},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.47049999237060547},{"id":"https://openalex.org/C2776257435","wikidata":"https://www.wikidata.org/wiki/Q1576430","display_name":"Bandwidth (computing)","level":2,"score":0.43790000677108765},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.42879998683929443},{"id":"https://openalex.org/C187834632","wikidata":"https://www.wikidata.org/wiki/Q188804","display_name":"Factorization","level":2,"score":0.41600000858306885},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3984000086784363},{"id":"https://openalex.org/C164226766","wikidata":"https://www.wikidata.org/wiki/Q7293202","display_name":"Rank (graph theory)","level":2,"score":0.350600004196167},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.34119999408721924},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.31040000915527344},{"id":"https://openalex.org/C155281189","wikidata":"https://www.wikidata.org/wiki/Q3518150","display_name":"Tensor (intrinsic definition)","level":2,"score":0.30799999833106995},{"id":"https://openalex.org/C761482","wikidata":"https://www.wikidata.org/wiki/Q118093","display_name":"Transmission (telecommunications)","level":2,"score":0.3018999993801117},{"id":"https://openalex.org/C90199385","wikidata":"https://www.wikidata.org/wiki/Q6692777","display_name":"Low-rank approximation","level":3,"score":0.2890999913215637},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2858999967575073},{"id":"https://openalex.org/C13481523","wikidata":"https://www.wikidata.org/wiki/Q412438","display_name":"Image compression","level":4,"score":0.2797999978065491},{"id":"https://openalex.org/C106195933","wikidata":"https://www.wikidata.org/wiki/Q7847935","display_name":"Truncation (statistics)","level":2,"score":0.27869999408721924},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.27000001072883606},{"id":"https://openalex.org/C42355184","wikidata":"https://www.wikidata.org/wiki/Q1361088","display_name":"Matrix decomposition","level":3,"score":0.26840001344680786},{"id":"https://openalex.org/C52970973","wikidata":"https://www.wikidata.org/wiki/Q2497134","display_name":"Adaptive system","level":2,"score":0.260699987411499},{"id":"https://openalex.org/C152124472","wikidata":"https://www.wikidata.org/wiki/Q1204361","display_name":"Redundancy (engineering)","level":2,"score":0.2547000050544739},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.25110000371932983}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2602.02696","doi":"https://doi.org/10.48550/arxiv.2602.02696","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.02696","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2602.02696","doi":"https://doi.org/10.48550/arxiv.2602.02696","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.02696","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"Preprint"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"The":[0,101],"expanding":[1],"scale":[2],"of":[3,43,70,124],"neural":[4],"networks":[5],"poses":[6],"a":[7,55],"major":[8],"challenge":[9],"for":[10,60,79,116],"distributed":[11],"machine":[12],"learning,":[13],"particularly":[14],"under":[15],"limited":[16],"communication":[17,38],"resources.":[18],"While":[19],"split":[20],"learning":[21],"(SL)":[22],"alleviates":[23],"client":[24],"computational":[25],"burden":[26],"by":[27],"distributing":[28],"model":[29],"layers":[30],"between":[31],"clients":[32],"and":[33,46],"server,":[34],"it":[35],"incurs":[36],"substantial":[37],"overhead":[39],"from":[40],"frequent":[41],"transmission":[42],"intermediate":[44],"activations":[45],"gradients.":[47],"To":[48],"tackle":[49],"this":[50],"issue,":[51],"we":[52],"propose":[53],"NSC-SL,":[54],"bandwidth-aware":[56],"adaptive":[57],"compression":[58,109],"algorithm":[59],"communication-efficient":[61],"SL.":[62],"NSC-SL":[63,85,105],"first":[64],"dynamically":[65],"determines":[66],"the":[67,75,121],"optimal":[68],"rank":[69],"low-rank":[71],"approximation":[72],"based":[73],"on":[74],"singular":[76],"value":[77],"distribution":[78],"adapting":[80],"real-time":[81],"bandwidth":[82],"constraints.":[83],"Then,":[84],"performs":[86],"error-compensated":[87],"tensor":[88],"factorization":[89],"using":[90],"alternating":[91],"orthogonal":[92],"iteration":[93],"with":[94],"residual":[95],"feedback,":[96],"effectively":[97],"minimizing":[98],"truncation":[99],"loss.":[100],"collaborative":[102],"mechanisms":[103],"enable":[104],"to":[106],"achieve":[107],"high":[108],"ratios":[110],"while":[111],"preserving":[112],"semantic-rich":[113],"information":[114],"essential":[115],"convergence.":[117],"Extensive":[118],"experiments":[119],"demonstrate":[120],"superb":[122],"performance":[123],"NSC-SL.":[125]},"counts_by_year":[],"updated_date":"2026-09-26T08:19:03.415552","created_date":"2026-02-06T00:00:00"}