{"id":"https://openalex.org/W2962705013","doi":"https://doi.org/10.1145/3234944.3234978","title":"MQGrad","display_name":"MQGrad","publication_year":2018,"publication_date":"2018-09-10","ids":{"openalex":"https://openalex.org/W2962705013","doi":"https://doi.org/10.1145/3234944.3234978","mag":"2962705013"},"language":"en","primary_location":{"id":"doi:10.1145/3234944.3234978","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3234944.3234978","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2018 ACM SIGIR International Conference on Theory of Information Retrieval","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/A5023311322","display_name":"Guoxin Cui","orcid":null},"institutions":[{"id":"https://openalex.org/I4210165038","display_name":"University of Chinese Academy of Sciences","ror":"https://ror.org/05qbk4x57","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165038"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Guoxin Cui","raw_affiliation_strings":["University of Chinese Academy of Sciences & Chinese Academy of Sciences, Beijing, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Chinese Academy of Sciences & Chinese Academy of Sciences, Beijing, China","institution_ids":["https://openalex.org/I4210165038"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5020766468","display_name":"Jun Xu","orcid":"https://orcid.org/0000-0001-7170-111X"},"institutions":[{"id":"https://openalex.org/I4210165038","display_name":"University of Chinese Academy of Sciences","ror":"https://ror.org/05qbk4x57","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165038"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jun Xu","raw_affiliation_strings":["University of Chinese Academy of Sciences & Chinese Academy of Sciences, Beijing, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Chinese Academy of Sciences & Chinese Academy of Sciences, Beijing, China","institution_ids":["https://openalex.org/I4210165038"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101427959","display_name":"Wei Zeng","orcid":"https://orcid.org/0009-0006-4437-9042"},"institutions":[{"id":"https://openalex.org/I4210165038","display_name":"University of Chinese Academy of Sciences","ror":"https://ror.org/05qbk4x57","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165038"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wei Zeng","raw_affiliation_strings":["University of Chinese Academy of Sciences & Chinese Academy of Sciences, Beijing, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Chinese Academy of Sciences & Chinese Academy of Sciences, Beijing, China","institution_ids":["https://openalex.org/I4210165038"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101616866","display_name":"Yanyan Lan","orcid":"https://orcid.org/0000-0002-7811-3262"},"institutions":[{"id":"https://openalex.org/I4210165038","display_name":"University of Chinese Academy of Sciences","ror":"https://ror.org/05qbk4x57","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165038"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yanyan Lan","raw_affiliation_strings":["University of Chinese Academy of Sciences & Chinese Academy of Sciences, Beijing, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Chinese Academy of Sciences & Chinese Academy of Sciences, Beijing, China","institution_ids":["https://openalex.org/I4210165038"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5088621320","display_name":"Jiafeng Guo","orcid":"https://orcid.org/0000-0002-9509-8674"},"institutions":[{"id":"https://openalex.org/I4210165038","display_name":"University of Chinese Academy of Sciences","ror":"https://ror.org/05qbk4x57","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165038"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jiafeng Guo","raw_affiliation_strings":["University of Chinese Academy of Sciences & Chinese Academy of Sciences, Beijing, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Chinese Academy of Sciences & Chinese Academy of Sciences, Beijing, China","institution_ids":["https://openalex.org/I4210165038"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5029998682","display_name":"Xueqi Cheng","orcid":"https://orcid.org/0000-0002-5201-8195"},"institutions":[{"id":"https://openalex.org/I4210165038","display_name":"University of Chinese Academy of Sciences","ror":"https://ror.org/05qbk4x57","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165038"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xueqi Cheng","raw_affiliation_strings":["University of Chinese Academy of Sciences & Chinese Academy of Sciences, Beijing, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Chinese Academy of Sciences & Chinese Academy of Sciences, Beijing, China","institution_ids":["https://openalex.org/I4210165038"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4210165038"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":13,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"83","last_page":"90"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9998000264167786,"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":0.9998000264167786,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9994999766349792,"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":0.9986000061035156,"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/computer-science","display_name":"Computer science","score":0.7910839319229126},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.7402181625366211},{"id":"https://openalex.org/keywords/markov-decision-process","display_name":"Markov decision process","score":0.6643606424331665},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.5544970631599426},{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.49810051918029785},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.48906761407852173},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.4368533492088318},{"id":"https://openalex.org/keywords/markov-process","display_name":"Markov process","score":0.4206186532974243},{"id":"https://openalex.org/keywords/server","display_name":"Server","score":0.4189959168434143},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.41093194484710693},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.3782789707183838},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.3506695628166199},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.14472749829292297},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.08789041638374329}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7910839319229126},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.7402181625366211},{"id":"https://openalex.org/C106189395","wikidata":"https://www.wikidata.org/wiki/Q176789","display_name":"Markov decision process","level":3,"score":0.6643606424331665},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.5544970631599426},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.49810051918029785},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.48906761407852173},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4368533492088318},{"id":"https://openalex.org/C159886148","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov process","level":2,"score":0.4206186532974243},{"id":"https://openalex.org/C93996380","wikidata":"https://www.wikidata.org/wiki/Q44127","display_name":"Server","level":2,"score":0.4189959168434143},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.41093194484710693},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3782789707183838},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.3506695628166199},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.14472749829292297},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.08789041638374329},{"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/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.0},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3234944.3234978","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3234944.3234978","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2018 ACM SIGIR International Conference on Theory of Information Retrieval","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Peace, Justice and strong institutions","score":0.6299999952316284,"id":"https://metadata.un.org/sdg/16"}],"awards":[{"id":"https://openalex.org/G6620960268","display_name":null,"funder_award_id":"61773362, 61425016, 61472401, 61722211, 20180290","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":21,"referenced_works":["https://openalex.org/W41554520","https://openalex.org/W1515851193","https://openalex.org/W1561921554","https://openalex.org/W1575599472","https://openalex.org/W1757796397","https://openalex.org/W2083842231","https://openalex.org/W2114766824","https://openalex.org/W2120972216","https://openalex.org/W2125389748","https://openalex.org/W2132737349","https://openalex.org/W2168231600","https://openalex.org/W2312609093","https://openalex.org/W2407022425","https://openalex.org/W2465345967","https://openalex.org/W2531786980","https://openalex.org/W2553303224","https://openalex.org/W2617766261","https://openalex.org/W2618530766","https://openalex.org/W2950656546","https://openalex.org/W2964299589","https://openalex.org/W3118608800"],"related_works":["https://openalex.org/W2595172197","https://openalex.org/W2084856301","https://openalex.org/W2127970246","https://openalex.org/W2885125400","https://openalex.org/W1989889224","https://openalex.org/W4382618745","https://openalex.org/W2315999538","https://openalex.org/W187740018","https://openalex.org/W2162286586","https://openalex.org/W4255368532"],"abstract_inverted_index":{"One":[0,53],"of":[1,63,71,126,146,158,185,194,209,225,245],"the":[2,18,27,31,36,50,61,67,76,81,94,102,107,111,120,144,156,170,174,178,183,186,191,202,207,210,215,221,229,243],"most":[3],"significant":[4],"bottleneck":[5],"in":[6,56,163],"training":[7,37,114,192],"large":[8,247],"scale":[9,248],"machine":[10,196],"learning":[11,197,244],"models":[12],"on":[13,214,234],"parameter":[14],"server":[15],"(PS)":[16],"is":[17,59],"communication":[19,51,77,95],"overhead,":[20],"because":[21],"it":[22],"needs":[23],"to":[24,48,106,152,206,219,227],"frequently":[25],"exchange":[26],"model":[28,98],"gradients":[29,121],"between":[30],"workers":[32],"and":[33,84,97],"servers":[34],"during":[35],"iterations.":[38,115],"Gradient":[39],"quantization":[40,58,89,141,159],"has":[41],"been":[42],"proposed":[43],"as":[44,153,161],"an":[45],"effective":[46],"approach":[47],"reducing":[49],"volume.":[52],"key":[54],"issue":[55],"gradient":[57,82],"setting":[60],"number":[62,70,103,125],"bits":[64,72,104,160],"for":[65],"quantizing":[66],"gradients.":[68,230],"Small":[69],"can":[73,241],"significantly":[74],"reduce":[75],"overhead":[78,96],"while":[79,252],"hurts":[80],"accuracies,":[83],"vise":[85],"versa.":[86],"An":[87],"ideal":[88],"method":[90,142],"would":[91],"dynamically":[92],"balance":[93],"accuracy,":[99],"through":[100],"adjusting":[101],"according":[105,205],"knowledge":[108],"learned":[109],"from":[110,177],"immediate":[112],"past":[113,179],"Existing":[116],"methods,":[117],"however,":[118],"quantize":[119,228],"either":[122],"with":[123,129],"fixed":[124],"bits,":[127],"or":[128],"predefined":[130],"heuristic":[131],"rules.":[132],"In":[133],"this":[134],"paper":[135],"we":[136],"propose":[137],"a":[138,164,195,235,246],"novel":[139],"adaptive":[140],"within":[143],"framework":[145],"reinforcement":[147],"learning.":[148],"The":[149],"method,":[150],"referred":[151],"MQGrad,":[154],"formalizes":[155],"selection":[157],"actions":[162,223],"Markov":[165],"decision":[166],"process":[167],"(MDP)":[168],"where":[169],"MDP":[171,203,217],"states":[172],"records":[173],"information":[175],"collected":[176],"optimization":[180],"iterations":[181,193],"(e.g.,":[182],"sequence":[184],"loss":[187,211],"function":[188],"values).":[189],"During":[190],"algorithm,":[198],"MQGrad":[199,240],"continuously":[200],"updates":[201],"state":[204],"changes":[208],"function.":[212],"Based":[213],"information,":[216],"learns":[218],"select":[220],"optimal":[222],"(number":[224],"bits)":[226],"Experimental":[231],"results":[232],"based":[233],"benchmark":[236],"dataset":[237],"showed":[238],"that":[239],"accelerate":[242],"deep":[249],"neural":[250],"network":[251],"keeping":[253],"its":[254],"prediction":[255],"accuracies.":[256]},"counts_by_year":[{"year":2025,"cited_by_count":3},{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":2},{"year":2022,"cited_by_count":2},{"year":2021,"cited_by_count":3},{"year":2020,"cited_by_count":2}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2019-07-30T00:00:00"}