{"id":"https://openalex.org/W2964279778","doi":"https://doi.org/10.5244/c.31.189","title":"Learning Accurate Low-Bit Deep Neural Networks with Stochastic Quantization","display_name":"Learning Accurate Low-Bit Deep Neural Networks with Stochastic Quantization","alternate_titles":[],"publication_year":2017,"publication_date":"2017-01-01","ids":{"openalex":"https://openalex.org/W2964279778","doi":"https://doi.org/10.5244/c.31.189","mag":"2964279778"},"language":"en","primary_location":{"id":"doi:10.5244/c.31.189","doi":"https://doi.org/10.5244/c.31.189","is_oa":false,"landing_page_url":"https://doi.org/10.5244/c.31.189","pdf_url":null,"source":{"id":"https://openalex.org/S7407088578","display_name":"British Machine Vision Conference (BMVC)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Procedings of the British Machine Vision Conference 2017","raw_type":"proceedings-article"},"parent_works":[],"lineage":["https://openalex.org/W2964279778"],"child_works_count":0,"type":"conference-paper","is_traditional":true,"indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/1708.01001","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5068755794","display_name":"Yinpeng Dong","orcid":"https://orcid.org/0000-0003-1299-683X","observed_orcids":["https://orcid.org/0000-0003-1299-683X"]},"institutions":[],"countries":[],"is_corresponding":false,"contributor_roles":[],"raw_author_name":"Yinpeng Dong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100368377","display_name":"Jianguo Li","orcid":"https://orcid.org/0000-0002-8645-0680","observed_orcids":["https://orcid.org/0000-0002-8645-0680"]},"institutions":[],"countries":[],"is_corresponding":false,"contributor_roles":[],"raw_author_name":"Jianguo Li","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5083225492","display_name":"Renkun Ni","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"contributor_roles":[],"raw_author_name":"Renkun Ni","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":1.7653,"has_fulltext":true,"cited_by_count":71,"citation_normalized_percentile":{"value":0.92526803,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":95,"max":99},"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_pulled":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9916999936103821,"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":0.9916999936103821,"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/T10363","display_name":"Low-power high-performance VLSI design","score":0.001500000013038516,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.0010000000474974513,"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/stochastic-quantization","display_name":"stochastic quantization","score":0.9929999709129333},{"id":"https://openalex.org/keywords/quantization-error","display_name":"quantization error","score":0.7889999747276306},{"id":"https://openalex.org/keywords/gradient-quantization","display_name":"gradient quantization","score":0.5260000228881836},{"id":"https://openalex.org/keywords/accuracy-preservation","display_name":"accuracy preservation","score":0.4189999997615814},{"id":"https://openalex.org/keywords/mixed-precision-training","display_name":"mixed-precision training","score":0.6489999890327454},{"id":"https://openalex.org/keywords/embedded-machine-learning","display_name":"embedded machine learning","score":0.5709999799728394}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7174043655395508},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.6220365762710571},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.6144514083862305},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.5361087322235107},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5116744041442871},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.45636409521102905},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.24664655327796936}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.5244/c.31.189","doi":"https://doi.org/10.5244/c.31.189","is_oa":false,"landing_page_url":"https://doi.org/10.5244/c.31.189","pdf_url":null,"source":{"id":"https://openalex.org/S7407088578","display_name":"British Machine Vision Conference (BMVC)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Procedings of the British Machine Vision Conference 2017","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:1708.01001","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/1708.01001","pdf_url":"https://arxiv.org/pdf/1708.01001","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.1708.01001","doi":"https://doi.org/10.48550/arxiv.1708.01001","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.1708.01001","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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:1708.01001","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/1708.01001","pdf_url":"https://arxiv.org/pdf/1708.01001","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G1919465925","display_name":null,"funder_award_id":"61621136008","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G2197472058","display_name":null,"funder_award_id":"2013CB329403","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G368795326","display_name":null,"funder_award_id":"61620106010","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G7373123958","display_name":null,"funder_award_id":"2013CB329403","funder_id":"https://openalex.org/F4320335777","funder_display_name":"National Key Research and Development Program of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320335777","display_name":"National Key Research and Development Program of China","ror":null}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W2964279778.pdf","grobid_xml":"https://content.openalex.org/works/W2964279778.grobid-xml"},"referenced_works_count":24,"referenced_works":["https://openalex.org/W639708223","https://openalex.org/W1686810756","https://openalex.org/W1845051632","https://openalex.org/W1902934009","https://openalex.org/W1923697677","https://openalex.org/W2095705004","https://openalex.org/W2102605133","https://openalex.org/W2117539524","https://openalex.org/W2155893237","https://openalex.org/W2194775991","https://openalex.org/W2291160084","https://openalex.org/W2300242332","https://openalex.org/W2319920447","https://openalex.org/W2331143823","https://openalex.org/W2395611524","https://openalex.org/W2405920868","https://openalex.org/W2469490737","https://openalex.org/W2524428287","https://openalex.org/W2529546376","https://openalex.org/W2560017826","https://openalex.org/W2593245696","https://openalex.org/W2919115771","https://openalex.org/W3118608800","https://openalex.org/W7098642242"],"related_works":["https://openalex.org/W2919311874","https://openalex.org/W3137025867","https://openalex.org/W3006769269","https://openalex.org/W3035282660","https://openalex.org/W4312454178","https://openalex.org/W3194014396","https://openalex.org/W3160434691","https://openalex.org/W4386822190","https://openalex.org/W2982997133","https://openalex.org/W2798742790"],"abstract_inverted_index":{"Low-bit":[0],"deep":[1],"neural":[2],"networks":[3],"(DNNs)":[4],"become":[5],"critical":[6],"for":[7,36,70,76,158,172],"embedded":[8],"applications":[9],"due":[10,44],"to":[11,45,80,99,107],"their":[12],"low":[13],"storage":[14],"requirement":[15],"and":[16,86,121,153,167,179],"computing":[17],"efficiency.":[18],"However,":[19],"they":[20],"suffer":[21],"much":[22],"from":[23],"the":[24,31,46,53,108,113,140,150,170],"non-negligible":[25],"accuracy":[26,90,157,171],"drop.":[27,91],"This":[28,145],"paper":[29],"proposes":[30],"stochastic":[32,103],"quantization":[33,65,109,151],"(SQ)":[34],"algorithm":[35],"learning":[37],"accurate":[38],"low-bit":[39,57,100,159,174],"DNNs.":[40,160],"The":[41,64,119,133],"motivation":[42],"is":[43,136,143],"following":[47],"observation.":[48],"Existing":[49],"training":[50],"algorithms":[51],"approximate":[52],"real-valued":[54],"elements/filters":[55,98],"with":[56,101,117,126],"representation":[58],"all":[59],"together":[60],"in":[61,130],"each":[62,131],"iteration.":[63,132],"errors":[66],"may":[67],"be":[68],"small":[69],"some":[71],"elements/filters,":[72],"while":[73,111],"are":[74,124],"remarkable":[75],"others,":[77],"which":[78],"lead":[79],"inappropriate":[81],"gradient":[82],"direction":[83],"during":[84],"training,":[85],"thus":[87,154],"bring":[88],"notable":[89],"Instead,":[92],"SQ":[93,134,164],"quantizes":[94],"a":[95,102],"portion":[96,115],"of":[97],"probability":[104],"inversely":[105],"proportional":[106],"error,":[110],"keeping":[112],"other":[114],"unchanged":[116],"full-precision.":[118],"quantized":[120],"full-precision":[122],"portions":[123],"updated":[125],"corresponding":[127],"gradients":[128],"separately":[129],"ratio":[135],"gradually":[137],"increased":[138],"until":[139],"whole":[141],"network":[142,181],"quantized.":[144],"procedure":[146],"can":[147,165],"greatly":[148],"compensate":[149],"error":[152],"yield":[155],"better":[156],"Experiments":[161],"show":[162],"that":[163],"consistently":[166],"significantly":[168],"improve":[169],"different":[173],"DNNs":[175],"on":[176],"various":[177,180],"datasets":[178],"structures.":[182]},"cited_by_api_url":"https://api.openalex.org/works?filter=cites:W2964279778","counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":3},{"year":2024,"cited_by_count":7},{"year":2023,"cited_by_count":11},{"year":2022,"cited_by_count":11},{"year":2021,"cited_by_count":11},{"year":2020,"cited_by_count":11},{"year":2019,"cited_by_count":7},{"year":2018,"cited_by_count":7}],"updated_date":"2026-11-21T10:27:49.728292","created_date":"2019-07-30T00:00:00"}
