{"id":"https://openalex.org/W3189787503","doi":"https://doi.org/10.1109/jsait.2021.3103494","title":"Compressing Gradients by Exploiting Temporal Correlation in Momentum-SGD","display_name":"Compressing Gradients by Exploiting Temporal Correlation in Momentum-SGD","publication_year":2021,"publication_date":"2021-08-11","ids":{"openalex":"https://openalex.org/W3189787503","doi":"https://doi.org/10.1109/jsait.2021.3103494","mag":"3189787503"},"language":"en","primary_location":{"id":"doi:10.1109/jsait.2021.3103494","is_oa":false,"landing_page_url":"https://doi.org/10.1109/jsait.2021.3103494","pdf_url":null,"source":{"id":"https://openalex.org/S4210211895","display_name":"IEEE Journal on Selected Areas in Information Theory","issn_l":"2641-8770","issn":["2641-8770"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Journal on Selected Areas in Information Theory","raw_type":"journal-article"},"type":"article","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2108.07827","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5076621312","display_name":"Tharindu Adikari","orcid":"https://orcid.org/0000-0002-7628-7568"},"institutions":[{"id":"https://openalex.org/I185261750","display_name":"University of Toronto","ror":"https://ror.org/03dbr7087","country_code":"CA","type":"education","lineage":["https://openalex.org/I185261750"]}],"countries":["CA"],"is_corresponding":true,"raw_author_name":"Tharindu B. Adikari","raw_affiliation_strings":["Electrical and Computer Engineering Department, University of Toronto, Toronto, Canada"],"affiliations":[{"raw_affiliation_string":"Electrical and Computer Engineering Department, University of Toronto, Toronto, Canada","institution_ids":["https://openalex.org/I185261750"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5029182703","display_name":"Stark C. Draper","orcid":"https://orcid.org/0000-0001-8100-5599"},"institutions":[{"id":"https://openalex.org/I185261750","display_name":"University of Toronto","ror":"https://ror.org/03dbr7087","country_code":"CA","type":"education","lineage":["https://openalex.org/I185261750"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Stark C. Draper","raw_affiliation_strings":["Electrical and Computer Engineering Department, University of Toronto, Toronto, Canada"],"affiliations":[{"raw_affiliation_string":"Electrical and Computer Engineering Department, University of Toronto, Toronto, Canada","institution_ids":["https://openalex.org/I185261750"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5076621312"],"corresponding_institution_ids":["https://openalex.org/I185261750"],"apc_list":null,"apc_paid":null,"fwci":0.2796,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.63620218,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":93,"max":99},"biblio":{"volume":"2","issue":"3","first_page":"970","last_page":"986"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9993000030517578,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9993000030517578,"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/T10500","display_name":"Sparse and Compressive Sensing Techniques","score":0.9988999962806702,"subfield":{"id":"https://openalex.org/subfields/2206","display_name":"Computational Mechanics"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11321","display_name":"Error Correcting Code Techniques","score":0.998199999332428,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/bottleneck","display_name":"Bottleneck","score":0.7039743661880493},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6936591863632202},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.6106076836585999},{"id":"https://openalex.org/keywords/rate-of-convergence","display_name":"Rate of convergence","score":0.5775174498558044},{"id":"https://openalex.org/keywords/convergence","display_name":"Convergence (economics)","score":0.5188435316085815},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.5037593245506287},{"id":"https://openalex.org/keywords/momentum","display_name":"Momentum (technical analysis)","score":0.4679728150367737},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.4633162021636963},{"id":"https://openalex.org/keywords/information-bottleneck-method","display_name":"Information bottleneck method","score":0.4582771956920624},{"id":"https://openalex.org/keywords/compression","display_name":"Compression (physics)","score":0.4294254183769226},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.4246354103088379},{"id":"https://openalex.org/keywords/norm","display_name":"Norm (philosophy)","score":0.4171103239059448},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.24601203203201294},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.1952769160270691},{"id":"https://openalex.org/keywords/telecommunications","display_name":"Telecommunications","score":0.09093287587165833}],"concepts":[{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.7039743661880493},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6936591863632202},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.6106076836585999},{"id":"https://openalex.org/C57869625","wikidata":"https://www.wikidata.org/wiki/Q1783502","display_name":"Rate of convergence","level":3,"score":0.5775174498558044},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.5188435316085815},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.5037593245506287},{"id":"https://openalex.org/C60718061","wikidata":"https://www.wikidata.org/wiki/Q1414747","display_name":"Momentum (technical analysis)","level":2,"score":0.4679728150367737},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.4633162021636963},{"id":"https://openalex.org/C60008888","wikidata":"https://www.wikidata.org/wiki/Q6031013","display_name":"Information bottleneck method","level":3,"score":0.4582771956920624},{"id":"https://openalex.org/C180016635","wikidata":"https://www.wikidata.org/wiki/Q2712821","display_name":"Compression (physics)","level":2,"score":0.4294254183769226},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.4246354103088379},{"id":"https://openalex.org/C191795146","wikidata":"https://www.wikidata.org/wiki/Q3878446","display_name":"Norm (philosophy)","level":2,"score":0.4171103239059448},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.24601203203201294},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.1952769160270691},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.09093287587165833},{"id":"https://openalex.org/C152139883","wikidata":"https://www.wikidata.org/wiki/Q252973","display_name":"Mutual information","level":2,"score":0.0},{"id":"https://openalex.org/C192562407","wikidata":"https://www.wikidata.org/wiki/Q228736","display_name":"Materials science","level":0,"score":0.0},{"id":"https://openalex.org/C10138342","wikidata":"https://www.wikidata.org/wiki/Q43015","display_name":"Finance","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C159985019","wikidata":"https://www.wikidata.org/wiki/Q181790","display_name":"Composite material","level":1,"score":0.0},{"id":"https://openalex.org/C199539241","wikidata":"https://www.wikidata.org/wiki/Q7748","display_name":"Law","level":1,"score":0.0},{"id":"https://openalex.org/C127162648","wikidata":"https://www.wikidata.org/wiki/Q16858953","display_name":"Channel (broadcasting)","level":2,"score":0.0},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.0},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.0},{"id":"https://openalex.org/C17744445","wikidata":"https://www.wikidata.org/wiki/Q36442","display_name":"Political science","level":0,"score":0.0},{"id":"https://openalex.org/C50522688","wikidata":"https://www.wikidata.org/wiki/Q189833","display_name":"Economic growth","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/jsait.2021.3103494","is_oa":false,"landing_page_url":"https://doi.org/10.1109/jsait.2021.3103494","pdf_url":null,"source":{"id":"https://openalex.org/S4210211895","display_name":"IEEE Journal on Selected Areas in Information Theory","issn_l":"2641-8770","issn":["2641-8770"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Journal on Selected Areas in Information Theory","raw_type":"journal-article"},{"id":"pmh:oai:arXiv.org:2108.07827","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2108.07827","pdf_url":"https://arxiv.org/pdf/2108.07827","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2108.07827","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2108.07827","pdf_url":"https://arxiv.org/pdf/2108.07827","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320322183","display_name":"Huawei Technologies","ror":"https://ror.org/00cmhce21"},{"id":"https://openalex.org/F4320334593","display_name":"Natural Sciences and Engineering Research Council of Canada","ror":"https://ror.org/01h531d29"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":55,"referenced_works":["https://openalex.org/W1533861849","https://openalex.org/W1766888123","https://openalex.org/W2030345669","https://openalex.org/W2162256416","https://openalex.org/W2401231614","https://openalex.org/W2405578611","https://openalex.org/W2407022425","https://openalex.org/W2504507112","https://openalex.org/W2563343794","https://openalex.org/W2606891064","https://openalex.org/W2617766261","https://openalex.org/W2741269719","https://openalex.org/W2741430497","https://openalex.org/W2766140019","https://openalex.org/W2769644379","https://openalex.org/W2774000609","https://openalex.org/W2787998955","https://openalex.org/W2805997383","https://openalex.org/W2890924858","https://openalex.org/W2896457183","https://openalex.org/W2911863041","https://openalex.org/W2945336802","https://openalex.org/W2950826569","https://openalex.org/W2963112712","https://openalex.org/W2963341956","https://openalex.org/W2963540381","https://openalex.org/W2963664311","https://openalex.org/W2963803379","https://openalex.org/W2964004663","https://openalex.org/W2964137095","https://openalex.org/W2964163156","https://openalex.org/W2964267428","https://openalex.org/W2971342441","https://openalex.org/W2978015420","https://openalex.org/W2990352720","https://openalex.org/W3101036738","https://openalex.org/W3160501684","https://openalex.org/W4230874317","https://openalex.org/W4297685247","https://openalex.org/W4297737050","https://openalex.org/W4301239768","https://openalex.org/W6631943919","https://openalex.org/W6713835734","https://openalex.org/W6738460352","https://openalex.org/W6741986022","https://openalex.org/W6745723224","https://openalex.org/W6746200960","https://openalex.org/W6746839373","https://openalex.org/W6748019269","https://openalex.org/W6751646691","https://openalex.org/W6752012617","https://openalex.org/W6754416507","https://openalex.org/W6758358981","https://openalex.org/W6762421227","https://openalex.org/W6795414890"],"related_works":["https://openalex.org/W2622284819","https://openalex.org/W1504394672","https://openalex.org/W3089381707","https://openalex.org/W4285254085","https://openalex.org/W3034190530","https://openalex.org/W2741297526","https://openalex.org/W4295728955","https://openalex.org/W3129794609","https://openalex.org/W2949033103","https://openalex.org/W2304083841"],"abstract_inverted_index":{"An":[0,54],"increasing":[1],"bottleneck":[2],"in":[3,50,85,105,115,163],"decentralized":[4],"optimization":[5],"is":[6,18,27,57,63,127],"communication.":[7],"Bigger":[8],"models":[9],"and":[10,20,77,89,165],"growing":[11],"datasets":[12],"mean":[13],"that":[14,21,48,81,98,141],"decentralization":[15],"of":[16,24,69,108,123,179],"computation":[17,116],"important":[19,55,155],"the":[22,39,45,66,94,106,121,132,150,177,191],"amount":[23],"information":[25],"exchanged":[26],"quickly":[28],"growing.":[29],"While":[30],"compression":[31,79,126],"techniques":[32],"have":[33],"been":[34],"introduced":[35],"to":[36,149],"cope":[37],"with":[38,88,93,129],"latter,":[40],"none":[41],"has":[42],"considered":[43],"leveraging":[44],"temporal":[46,61,83],"correlations":[47],"exist":[49],"consecutive":[51],"vector":[52],"updates.":[53],"example":[56],"distributed":[58],"momentum-SGD":[59],"where":[60],"correlation":[62,84],"enhanced":[64],"by":[65,186],"low-pass-filtering":[67],"effect":[68],"applying":[70],"momentum.":[71],"In":[72,131,152,172],"this":[73,173],"paper":[74,174],"we":[75,175],"design":[76],"analyze":[78,120],"methods":[80,101],"exploit":[82],"systems":[86],"both":[87],"without":[90],"error-feedback.":[91,130],"Experiments":[92],"ImageNet":[95],"dataset":[96],"demonstrate":[97],"our":[99],"proposed":[100],"offer":[102],"significant":[103],"reduction":[104],"rate":[107],"communication":[109],"at":[110],"only":[111,138,162],"a":[112,168,188],"negligible":[113],"increase":[114],"complexity.":[117],"We":[118],"further":[119],"convergence":[122,134,178],"SGD":[124,180],"when":[125],"applied":[128],"literature,":[133],"guarantees":[135],"are":[136],"developed":[137],"for":[139,146,190],"compressors":[140],"provide":[142,160,167],"error-bounds":[143,161],"point-wise,":[144],"i.e.,":[145],"each":[147],"input":[148],"compressor.":[151],"contrast,":[153],"many":[154],"codes":[156],"(e.g.,":[157],"rate-distortion":[158],"codes)":[159],"expectation":[164],"thus":[166],"more":[169],"general":[170],"guarantee.":[171],"prove":[176],"under":[181],"an":[182],"expected":[183],"error":[184],"assumption":[185],"establishing":[187],"bound":[189],"minimum":[192],"gradient":[193],"norm.":[194]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2021,"cited_by_count":2}],"updated_date":"2026-03-10T16:38:18.471706","created_date":"2025-10-10T00:00:00"}