{"id":"https://openalex.org/W2905549570","doi":"https://doi.org/10.1609/aaai.v33i01.33014173","title":"Communication-Efficient Stochastic Gradient MCMC for Neural Networks","display_name":"Communication-Efficient Stochastic Gradient MCMC for Neural Networks","publication_year":2019,"publication_date":"2019-07-17","ids":{"openalex":"https://openalex.org/W2905549570","doi":"https://doi.org/10.1609/aaai.v33i01.33014173","mag":"2905549570"},"language":"en","primary_location":{"id":"doi:10.1609/aaai.v33i01.33014173","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v33i01.33014173","pdf_url":"https://ojs.aaai.org/index.php/AAAI/article/download/5190/5057","source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://ojs.aaai.org/index.php/AAAI/article/download/5190/5057","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5107893340","display_name":"Chunyuan Li","orcid":null},"institutions":[{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Chunyuan Li","raw_affiliation_strings":["Microsoft Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research","institution_ids":["https://openalex.org/I4210164937"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102745649","display_name":"Changyou Chen","orcid":"https://orcid.org/0000-0002-3230-2770"},"institutions":[{"id":"https://openalex.org/I63190737","display_name":"University at Buffalo, State University of New York","ror":"https://ror.org/01y64my43","country_code":"US","type":"education","lineage":["https://openalex.org/I63190737"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Changyou Chen","raw_affiliation_strings":["State University of New York at Buffalo"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State University of New York at Buffalo","institution_ids":["https://openalex.org/I63190737"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5023417820","display_name":"Yunchen Pu","orcid":null},"institutions":[{"id":"https://openalex.org/I2252078561","display_name":"Meta (Israel)","ror":"https://ror.org/02388em19","country_code":"IL","type":"company","lineage":["https://openalex.org/I2252078561","https://openalex.org/I4210114444"]}],"countries":["IL"],"is_corresponding":false,"raw_author_name":"Yunchen Pu","raw_affiliation_strings":["Facebook"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Facebook","institution_ids":["https://openalex.org/I2252078561"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5056639842","display_name":"Ricardo Henao","orcid":"https://orcid.org/0000-0003-4980-845X"},"institutions":[{"id":"https://openalex.org/I170897317","display_name":"Duke University","ror":"https://ror.org/00py81415","country_code":"US","type":"education","lineage":["https://openalex.org/I170897317"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ricardo Henao","raw_affiliation_strings":["Duke University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Duke University","institution_ids":["https://openalex.org/I170897317"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5016448581","display_name":"Lawrence Carin","orcid":"https://orcid.org/0000-0001-6277-7948"},"institutions":[{"id":"https://openalex.org/I170897317","display_name":"Duke University","ror":"https://ror.org/00py81415","country_code":"US","type":"education","lineage":["https://openalex.org/I170897317"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Lawrence Carin","raw_affiliation_strings":["Duke University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Duke University","institution_ids":["https://openalex.org/I170897317"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":10,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":"33","issue":"01","first_page":"4173","last_page":"4180"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9968000054359436,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9968000054359436,"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":0.9962999820709229,"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":0.9955999851226807,"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/markov-chain-monte-carlo","display_name":"Markov chain Monte Carlo","score":0.8143765330314636},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7523181438446045},{"id":"https://openalex.org/keywords/asynchronous-communication","display_name":"Asynchronous communication","score":0.6559005975723267},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.6249160766601562},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.5184580683708191},{"id":"https://openalex.org/keywords/consistency","display_name":"Consistency (knowledge bases)","score":0.5173439383506775},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.5135021805763245},{"id":"https://openalex.org/keywords/bayesian-probability","display_name":"Bayesian probability","score":0.47232311964035034},{"id":"https://openalex.org/keywords/sample","display_name":"Sample (material)","score":0.4476880729198456},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.4425833225250244},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4377751350402832},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.4152162969112396},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.3787415623664856},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.3246484398841858},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.11691311001777649},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.07871043682098389}],"concepts":[{"id":"https://openalex.org/C111350023","wikidata":"https://www.wikidata.org/wiki/Q1191869","display_name":"Markov chain Monte Carlo","level":3,"score":0.8143765330314636},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7523181438446045},{"id":"https://openalex.org/C151319957","wikidata":"https://www.wikidata.org/wiki/Q752739","display_name":"Asynchronous communication","level":2,"score":0.6559005975723267},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.6249160766601562},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.5184580683708191},{"id":"https://openalex.org/C2776436953","wikidata":"https://www.wikidata.org/wiki/Q5163215","display_name":"Consistency (knowledge bases)","level":2,"score":0.5173439383506775},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.5135021805763245},{"id":"https://openalex.org/C107673813","wikidata":"https://www.wikidata.org/wiki/Q812534","display_name":"Bayesian probability","level":2,"score":0.47232311964035034},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.4476880729198456},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.4425833225250244},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4377751350402832},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4152162969112396},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3787415623664856},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.3246484398841858},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.11691311001777649},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.07871043682098389},{"id":"https://openalex.org/C43617362","wikidata":"https://www.wikidata.org/wiki/Q170050","display_name":"Chromatography","level":1,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1609/aaai.v33i01.33014173","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v33i01.33014173","pdf_url":"https://ojs.aaai.org/index.php/AAAI/article/download/5190/5057","source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},{"id":"pmh:oai:ojs.aaai.org:article/5190","is_oa":false,"landing_page_url":"https://ojs.aaai.org/index.php/AAAI/article/view/5190","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"2159-5399","raw_type":"info:eu-repo/semantics/article"}],"best_oa_location":{"id":"doi:10.1609/aaai.v33i01.33014173","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v33i01.33014173","pdf_url":"https://ojs.aaai.org/index.php/AAAI/article/download/5190/5057","source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W2905549570.pdf","grobid_xml":"https://content.openalex.org/works/W2905549570.grobid-xml"},"referenced_works_count":61,"referenced_works":["https://openalex.org/W3401598","https://openalex.org/W104184427","https://openalex.org/W123701566","https://openalex.org/W582134693","https://openalex.org/W778657980","https://openalex.org/W1522301498","https://openalex.org/W1524172054","https://openalex.org/W1567512734","https://openalex.org/W1592410721","https://openalex.org/W1719489212","https://openalex.org/W1737237432","https://openalex.org/W1826234144","https://openalex.org/W1951216520","https://openalex.org/W1973794074","https://openalex.org/W2095705004","https://openalex.org/W2101893470","https://openalex.org/W2144193737","https://openalex.org/W2156387975","https://openalex.org/W2157331557","https://openalex.org/W2165652770","https://openalex.org/W2167433878","https://openalex.org/W2168231600","https://openalex.org/W2186210550","https://openalex.org/W2187089797","https://openalex.org/W2238987678","https://openalex.org/W2257113116","https://openalex.org/W2260756217","https://openalex.org/W2263340393","https://openalex.org/W2352456897","https://openalex.org/W2431605385","https://openalex.org/W2511837229","https://openalex.org/W2549291898","https://openalex.org/W2556547645","https://openalex.org/W2559659309","https://openalex.org/W2623491082","https://openalex.org/W2624543814","https://openalex.org/W2724169821","https://openalex.org/W2732240984","https://openalex.org/W2950942124","https://openalex.org/W2951266961","https://openalex.org/W2951595529","https://openalex.org/W2953263857","https://openalex.org/W2963050127","https://openalex.org/W2963545724","https://openalex.org/W2963804082","https://openalex.org/W2963903325","https://openalex.org/W2964043796","https://openalex.org/W2964059111","https://openalex.org/W6600144256","https://openalex.org/W6631190155","https://openalex.org/W6631656437","https://openalex.org/W6633735129","https://openalex.org/W6635477909","https://openalex.org/W6638803421","https://openalex.org/W6666761814","https://openalex.org/W6681286327","https://openalex.org/W6712435533","https://openalex.org/W6729119400","https://openalex.org/W6729629034","https://openalex.org/W6739193204","https://openalex.org/W6764214684"],"related_works":["https://openalex.org/W2116677773","https://openalex.org/W2151689585","https://openalex.org/W2380816257","https://openalex.org/W3087071515","https://openalex.org/W2155261584","https://openalex.org/W4283726152","https://openalex.org/W2584231425","https://openalex.org/W2042919702","https://openalex.org/W3093571331","https://openalex.org/W4283028824"],"abstract_inverted_index":{"Learning":[0],"probability":[1],"distributions":[2],"on":[3,99,115],"the":[4,54,68,71,90,100,121],"weights":[5],"of":[6,104],"neural":[7,37,117],"networks":[8,118],"has":[9],"recently":[10],"proven":[11],"beneficial":[12],"in":[13,36,60],"many":[14],"applications.":[15],"Bayesian":[16],"methods":[17],"such":[18],"as":[19],"Stochastic":[20],"Gradient":[21],"Markov":[22],"Chain":[23],"Monte":[24],"Carlo":[25],"(SG-MCMC)":[26],"offer":[27],"an":[28],"elegant":[29],"framework":[30],"to":[31,85,110,138,143],"reason":[32],"about":[33],"model":[34],"uncertainty":[35],"networks.":[38],"However,":[39],"these":[40],"advantages":[41],"usually":[42],"come":[43],"with":[44,154],"a":[45,96],"high":[46],"computational":[47],"cost.":[48],"We":[49,94],"propose":[50],"accelerating":[51],"SG-MCMC":[52],"under":[53],"masterworker":[55],"framework:":[56],"workers":[57],"asynchronously":[58],"and":[59,81,92,107],"parallel":[61],"share":[62],"responsibility":[63],"for":[64,150],"gradient":[65],"computations,":[66],"while":[67,129],"master":[69,91],"collects":[70],"final":[72],"samples.":[73],"To":[74],"reduce":[75,126],"communication":[76],"overhead,":[77],"two":[78],"protocols":[79],"(downpour":[80],"elastic)":[82],"are":[83],"developed":[84],"allow":[86],"periodic":[87],"interaction":[88],"between":[89],"workers.":[93],"provide":[95],"theoretical":[97],"analysis":[98],"finite-time":[101],"estimation":[102],"consistency":[103],"posterior":[105],"expectations,":[106],"establish":[108],"connections":[109],"sample":[111],"thinning.":[112],"Our":[113],"experiments":[114],"various":[116],"demonstrate":[119],"that":[120],"proposed":[122],"algorithms":[123],"can":[124],"greatly":[125],"training":[127],"time":[128],"achieving":[130],"comparable":[131],"(or":[132],"better)":[133],"test":[134],"accuracy/log-likelihood":[135],"levels,":[136],"relative":[137],"traditional":[139],"SG-MCMC.":[140],"When":[141],"applied":[142],"reinforcement":[144],"learning,":[145],"it":[146],"naturally":[147],"provides":[148],"exploration":[149],"asynchronous":[151],"policy":[152],"optimization,":[153],"encouraging":[155],"performance":[156],"improvement.":[157]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2021,"cited_by_count":2},{"year":2020,"cited_by_count":4},{"year":2019,"cited_by_count":2},{"year":2018,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}