{"id":"https://openalex.org/W4415708171","doi":"https://doi.org/10.1109/icme59968.2025.11209773","title":"Adaptive Gradient Quantization with Bit Allocation for Distributed Deep Learning","display_name":"Adaptive Gradient Quantization with Bit Allocation for Distributed Deep Learning","publication_year":2025,"publication_date":"2025-06-30","ids":{"openalex":"https://openalex.org/W4415708171","doi":"https://doi.org/10.1109/icme59968.2025.11209773"},"language":null,"primary_location":{"id":"doi:10.1109/icme59968.2025.11209773","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icme59968.2025.11209773","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE International Conference on Multimedia and Expo (ICME)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100318657","display_name":"Fei Gao","orcid":"https://orcid.org/0000-0001-8626-5509"},"institutions":[{"id":"https://openalex.org/I4210136793","display_name":"Peng Cheng Laboratory","ror":"https://ror.org/03qdqbt06","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210136793"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Fei Gao","raw_affiliation_strings":["Peng Cheng Laboratory,Shenzhen,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Peng Cheng Laboratory,Shenzhen,China","institution_ids":["https://openalex.org/I4210136793"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5026281370","display_name":"Xingyu Yan","orcid":"https://orcid.org/0000-0002-4621-3584"},"institutions":[{"id":"https://openalex.org/I4210100070","display_name":"Guangzhou Mechanical Engineering Research Institute (China)","ror":"https://ror.org/0114hcc46","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210100070"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xingyu Yan","raw_affiliation_strings":["China-Singapore International Joint Research Institute (CSIJRI),Guangzhou,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"China-Singapore International Joint Research Institute (CSIJRI),Guangzhou,China","institution_ids":["https://openalex.org/I4210100070"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101678706","display_name":"Jian Jin","orcid":"https://orcid.org/0000-0003-4250-1519"},"institutions":[{"id":"https://openalex.org/I172675005","display_name":"Nanyang Technological University","ror":"https://ror.org/02e7b5302","country_code":"SG","type":"education","lineage":["https://openalex.org/I172675005"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Jian Jin","raw_affiliation_strings":["Nanyang Technological University,Singapore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nanyang Technological University,Singapore","institution_ids":["https://openalex.org/I172675005"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5070884682","display_name":"Wenhan Yang","orcid":"https://orcid.org/0000-0002-1692-0069"},"institutions":[{"id":"https://openalex.org/I4210136793","display_name":"Peng Cheng Laboratory","ror":"https://ror.org/03qdqbt06","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210136793"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wenhan Yang","raw_affiliation_strings":["Peng Cheng Laboratory,Shenzhen,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Peng Cheng Laboratory,Shenzhen,China","institution_ids":["https://openalex.org/I4210136793"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Lingyu Duan","orcid":null},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lingyu Duan","raw_affiliation_strings":["Peking University,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Peking University,Beijing,China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100345083","display_name":"Zhuo Chen","orcid":"https://orcid.org/0000-0002-6590-3935"},"institutions":[{"id":"https://openalex.org/I4210136793","display_name":"Peng Cheng Laboratory","ror":"https://ror.org/03qdqbt06","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210136793"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhuo Chen","raw_affiliation_strings":["Peng Cheng Laboratory,Shenzhen,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Peng Cheng Laboratory,Shenzhen,China","institution_ids":["https://openalex.org/I4210136793"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.34314728,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"6"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.6281999945640564,"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":0.6281999945640564,"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/T10901","display_name":"Advanced Data Compression Techniques","score":0.09440000355243683,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.06669999659061432,"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/quantization","display_name":"Quantization (signal processing)","score":0.7641000151634216},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.5052000284194946},{"id":"https://openalex.org/keywords/bit","display_name":"Bit (key)","score":0.4551999866962433},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.4262999892234802},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.40529999136924744},{"id":"https://openalex.org/keywords/data-compression","display_name":"Data compression","score":0.3149999976158142},{"id":"https://openalex.org/keywords/gradient-descent","display_name":"Gradient descent","score":0.30889999866485596}],"concepts":[{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.7641000151634216},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7361000180244446},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.5231000185012817},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.5052000284194946},{"id":"https://openalex.org/C117011727","wikidata":"https://www.wikidata.org/wiki/Q1278488","display_name":"Bit (key)","level":2,"score":0.4551999866962433},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4262999892234802},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.40529999136924744},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3610000014305115},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.3149999976158142},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.30889999866485596},{"id":"https://openalex.org/C204323151","wikidata":"https://www.wikidata.org/wiki/Q905424","display_name":"Range (aeronautics)","level":2,"score":0.3043999969959259},{"id":"https://openalex.org/C117896860","wikidata":"https://www.wikidata.org/wiki/Q11376","display_name":"Acceleration","level":2,"score":0.29339998960494995},{"id":"https://openalex.org/C183383400","wikidata":"https://www.wikidata.org/wiki/Q2374485","display_name":"Bit field","level":3,"score":0.27320000529289246},{"id":"https://openalex.org/C51823790","wikidata":"https://www.wikidata.org/wiki/Q504353","display_name":"Greedy algorithm","level":2,"score":0.2721000015735626},{"id":"https://openalex.org/C56296756","wikidata":"https://www.wikidata.org/wiki/Q840922","display_name":"Bit error rate","level":3,"score":0.2635999917984009},{"id":"https://openalex.org/C13481523","wikidata":"https://www.wikidata.org/wiki/Q412438","display_name":"Image compression","level":4,"score":0.26350000500679016}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icme59968.2025.11209773","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icme59968.2025.11209773","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE International Conference on Multimedia and Expo (ICME)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":8,"referenced_works":["https://openalex.org/W2083842231","https://openalex.org/W2194775991","https://openalex.org/W2407022425","https://openalex.org/W2511584835","https://openalex.org/W2606891064","https://openalex.org/W2970102825","https://openalex.org/W2982479999","https://openalex.org/W3174633912"],"related_works":[],"abstract_inverted_index":{"Gradient":[0],"compression":[1,16,178],"plays":[2],"a":[3,43,57],"crucial":[4],"role":[5],"in":[6,10,35],"mitigating":[7],"communication":[8],"overhead":[9],"distributed":[11],"deep":[12,94],"learning.":[13],"Existing":[14],"gradient":[15,49,144,177],"methods":[17,142,170],"usually":[18],"employ":[19],"fix-bit":[20],"quantization":[21,50,54,157],"across":[22,127],"all":[23],"layers,":[24],"neglecting":[25],"the":[26,63,86,100,112,117,135],"varying":[27],"sensitivities":[28],"of":[29,67,137,148],"different":[30],"layers":[31],"to":[32,80,110],"compression,":[33],"resulting":[34],"suboptimal":[36],"performance.":[37,182],"In":[38],"this":[39],"paper,":[40],"we":[41,75,104],"introduce":[42],"layer-wise":[44],"bit":[45,59,73,88,101,114,140,168],"allocation":[46,89,102,141,169],"mechanism":[47],"for":[48,72,92,143],"that":[51],"minimizes":[52],"overall":[53,181],"error":[55],"within":[56],"specified":[58],"budget.":[60],"To":[61],"address":[62],"heavy":[64],"computational":[65,82],"load":[66],"conventional":[68],"greedy":[69],"search":[70],"approach":[71],"allocation,":[74],"develop":[76],"two":[77],"acceleration":[78,150],"techniques":[79,151],"reduce":[81],"overhead,":[83],"thereby":[84],"making":[85],"proposed":[87,139,149,167],"method":[90],"feasible":[91],"real-time":[93],"learning":[95],"training.":[96],"Specifically,":[97],"by":[98],"observing":[99],"statistics,":[103],"propose":[105],"Bit":[106,118],"Searching":[107],"Range":[108],"Optimization":[109],"narrow":[111],"available":[113],"options,":[115],"while":[116],"Pre-Assignment":[119],"selectively":[120],"bypasses":[121],"certain":[122],"searching":[123],"processes.":[124],"Experimental":[125],"results":[126],"various":[128],"neural":[129],"network":[130],"models":[131],"and":[132,162],"datasets":[133],"demonstrate":[134],"effectiveness":[136],"our":[138,166],"quantization.":[145],"The":[146],"combination":[147],"offers":[152],"an":[153],"advantageous":[154],"trade-off":[155],"among":[156],"error,":[158],"model":[159],"training":[160],"performance":[161],"time":[163],"consumption.":[164],"Moreover,":[165],"can":[171],"be":[172],"seamlessly":[173],"integrated":[174],"with":[175],"existing":[176],"approaches,":[179],"improving":[180]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-30T00:00:00"}