{"id":"https://openalex.org/W4386709680","doi":"https://doi.org/10.1145/3605573.3605580","title":"Output-Directed Dynamic Quantization for DNN Acceleration","display_name":"Output-Directed Dynamic Quantization for DNN Acceleration","publication_year":2023,"publication_date":"2023-08-07","ids":{"openalex":"https://openalex.org/W4386709680","doi":"https://doi.org/10.1145/3605573.3605580"},"language":"en","primary_location":{"id":"doi:10.1145/3605573.3605580","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3605573.3605580","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3605573.3605580","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 52nd International Conference on Parallel Processing","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3605573.3605580","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5090623049","display_name":"Beilei Jiang","orcid":"https://orcid.org/0009-0001-3985-333X"},"institutions":[{"id":"https://openalex.org/I123534392","display_name":"University of North Texas","ror":"https://ror.org/00v97ad02","country_code":"US","type":"education","lineage":["https://openalex.org/I123534392"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Beilei Jiang","raw_affiliation_strings":["University of North Texas, United States of America"],"raw_orcid":"https://orcid.org/0009-0001-3985-333X","affiliations":[{"raw_affiliation_string":"University of North Texas, United States of America","institution_ids":["https://openalex.org/I123534392"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003848547","display_name":"Xianwei Cheng","orcid":"https://orcid.org/0009-0001-3550-2647"},"institutions":[{"id":"https://openalex.org/I123534392","display_name":"University of North Texas","ror":"https://ror.org/00v97ad02","country_code":"US","type":"education","lineage":["https://openalex.org/I123534392"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xianwei Cheng","raw_affiliation_strings":["University of North Texas, United States of America"],"raw_orcid":"https://orcid.org/0009-0001-3550-2647","affiliations":[{"raw_affiliation_string":"University of North Texas, United States of America","institution_ids":["https://openalex.org/I123534392"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101512990","display_name":"Yuan Li","orcid":"https://orcid.org/0009-0006-9301-9369"},"institutions":[{"id":"https://openalex.org/I123534392","display_name":"University of North Texas","ror":"https://ror.org/00v97ad02","country_code":"US","type":"education","lineage":["https://openalex.org/I123534392"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yuan Li","raw_affiliation_strings":["University of North Texas, United States of America"],"raw_orcid":"https://orcid.org/0009-0006-9301-9369","affiliations":[{"raw_affiliation_string":"University of North Texas, United States of America","institution_ids":["https://openalex.org/I123534392"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5091390930","display_name":"Jocelyn Zhang","orcid":"https://orcid.org/0009-0008-9828-115X"},"institutions":[{"id":"https://openalex.org/I123534392","display_name":"University of North Texas","ror":"https://ror.org/00v97ad02","country_code":"US","type":"education","lineage":["https://openalex.org/I123534392"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jocelyn Zhang","raw_affiliation_strings":["University of North Texas, USA"],"raw_orcid":"https://orcid.org/0009-0008-9828-115X","affiliations":[{"raw_affiliation_string":"University of North Texas, USA","institution_ids":["https://openalex.org/I123534392"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5082758762","display_name":"Song Fu","orcid":"https://orcid.org/0000-0002-7705-0829"},"institutions":[{"id":"https://openalex.org/I123534392","display_name":"University of North Texas","ror":"https://ror.org/00v97ad02","country_code":"US","type":"education","lineage":["https://openalex.org/I123534392"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Song Fu","raw_affiliation_strings":["University of North Texas, United States of America"],"raw_orcid":"https://orcid.org/0000-0002-7705-0829","affiliations":[{"raw_affiliation_string":"University of North Texas, United States of America","institution_ids":["https://openalex.org/I123534392"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102751945","display_name":"Qing Yang","orcid":"https://orcid.org/0000-0003-3495-370X"},"institutions":[{"id":"https://openalex.org/I123534392","display_name":"University of North Texas","ror":"https://ror.org/00v97ad02","country_code":"US","type":"education","lineage":["https://openalex.org/I123534392"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Qing Yang","raw_affiliation_strings":["University of North Texas, United States of America"],"raw_orcid":"https://orcid.org/0000-0003-3495-370X","affiliations":[{"raw_affiliation_string":"University of North Texas, United States of America","institution_ids":["https://openalex.org/I123534392"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5105783836","display_name":"M. X. Liu","orcid":"https://orcid.org/0000-0002-5992-1221"},"institutions":[{"id":"https://openalex.org/I1343871089","display_name":"Los Alamos National Laboratory","ror":"https://ror.org/01e41cf67","country_code":"US","type":"facility","lineage":["https://openalex.org/I1330989302","https://openalex.org/I1343871089","https://openalex.org/I198811213","https://openalex.org/I4210120050"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Mingxiong Liu","raw_affiliation_strings":["Los Alamos National Laboratory, United States of America"],"raw_orcid":"https://orcid.org/0000-0002-5992-1221","affiliations":[{"raw_affiliation_string":"Los Alamos National Laboratory, United States of America","institution_ids":["https://openalex.org/I1343871089"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5087047310","display_name":"Alejandro Olvera","orcid":"https://orcid.org/0000-0001-9112-0734"},"institutions":[{"id":"https://openalex.org/I123534392","display_name":"University of North Texas","ror":"https://ror.org/00v97ad02","country_code":"US","type":"education","lineage":["https://openalex.org/I123534392"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Alejandro Olvera","raw_affiliation_strings":["University of North Texas, United States of America"],"raw_orcid":"https://orcid.org/0000-0001-9112-0734","affiliations":[{"raw_affiliation_string":"University of North Texas, United States of America","institution_ids":["https://openalex.org/I123534392"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.2426,"has_fulltext":true,"cited_by_count":2,"citation_normalized_percentile":{"value":0.57254125,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":96},"biblio":{"volume":null,"issue":null,"first_page":"645","last_page":"654"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9998999834060669,"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":0.9998999834060669,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.994700014591217,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9897000193595886,"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":0.8696152567863464},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7152366042137146},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.7149215936660767},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.5949482321739197},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.49670273065567017},{"id":"https://openalex.org/keywords/sensitivity","display_name":"Sensitivity (control systems)","score":0.4632936716079712},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.44163209199905396},{"id":"https://openalex.org/keywords/electronic-engineering","display_name":"Electronic engineering","score":0.24731457233428955},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.13023746013641357},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.11493709683418274},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.07812753319740295}],"concepts":[{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.8696152567863464},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7152366042137146},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.7149215936660767},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.5949482321739197},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.49670273065567017},{"id":"https://openalex.org/C21200559","wikidata":"https://www.wikidata.org/wiki/Q7451068","display_name":"Sensitivity (control systems)","level":2,"score":0.4632936716079712},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.44163209199905396},{"id":"https://openalex.org/C24326235","wikidata":"https://www.wikidata.org/wiki/Q126095","display_name":"Electronic engineering","level":1,"score":0.24731457233428955},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.13023746013641357},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.11493709683418274},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.07812753319740295}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3605573.3605580","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3605573.3605580","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3605573.3605580","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 52nd International Conference on Parallel Processing","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3605573.3605580","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3605573.3605580","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3605573.3605580","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 52nd International Conference on Parallel Processing","raw_type":"proceedings-article"},"sustainable_development_goals":[{"display_name":"Affordable and clean energy","score":0.9100000262260437,"id":"https://metadata.un.org/sdg/7"}],"awards":[{"id":"https://openalex.org/G1661319800","display_name":null,"funder_award_id":"CNS-1852134","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G1731975950","display_name":null,"funder_award_id":"OAC-2017564","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G1784947659","display_name":null,"funder_award_id":"CNS-2113805","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G1972538445","display_name":"CyberTraining: Implementation: Small: Collaborative and Integrated Training on Connected and Autonomous Vehicles Cyber Infrastructure","funder_award_id":"2017564","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G3561560200","display_name":"REU Site: Vehicular Edge Computing and Security: Research Experience for Undergraduates","funder_award_id":"1852134","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G4063260910","display_name":"EAGER: SaTC: Privacy-Preserving Convolutional Neural Network for Cooperative Perception in Vehicular Edge Systems","funder_award_id":"2037982","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G5793853297","display_name":null,"funder_award_id":"CNS-2231519","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G586731781","display_name":"IUCRC Phase I University of North Texas: Center for Electric, Connected and Autonomous Technologies for Mobility (eCAT)","funder_award_id":"2231519","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G6278895685","display_name":null,"funder_award_id":"CNS-1828105","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G6826730685","display_name":null,"funder_award_id":"ECCS-2010332","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G7515869040","display_name":null,"funder_award_id":"DUE-2225229","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G7550793203","display_name":null,"funder_award_id":"CNS-2037982","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G7868034680","display_name":"MRI Collaborative: Development of ESPRIT - Emerging systems' performance and energy evaluation instruments and testbench","funder_award_id":"1828105","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G8026271096","display_name":"Developing a High-Quality Academic Environment for Broadening Participation of Hispanic Students in Computing","funder_award_id":"2225229","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G8398141313","display_name":"Collaborative Research: Enabling Machine Learning based Cooperative Perception with mmWave Communication for Autonomous Vehicle Safety","funder_award_id":"2010332","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G8418168475","display_name":"IUCRC Planning Grant University of North Texas: Center for Electric, Connected and Autonomous Technologies for Mobility (eCAT)","funder_award_id":"2113805","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320337392","display_name":"Division of Electrical, Communications and Cyber Systems","ror":"https://ror.org/01krpsy48"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4386709680.pdf","grobid_xml":"https://content.openalex.org/works/W4386709680.grobid-xml"},"referenced_works_count":11,"referenced_works":["https://openalex.org/W1984541135","https://openalex.org/W2096645269","https://openalex.org/W2809541887","https://openalex.org/W2883920103","https://openalex.org/W2963367920","https://openalex.org/W2963446712","https://openalex.org/W2963749936","https://openalex.org/W3043504674","https://openalex.org/W3099971873","https://openalex.org/W3158020960","https://openalex.org/W4236965008"],"related_works":["https://openalex.org/W2058965144","https://openalex.org/W2164382479","https://openalex.org/W2146343568","https://openalex.org/W98480971","https://openalex.org/W2150291671","https://openalex.org/W2013643406","https://openalex.org/W2027972911","https://openalex.org/W2157978810","https://openalex.org/W2597809628","https://openalex.org/W3046370962"],"abstract_inverted_index":{"Quantization":[0],"is":[1,22,109,160],"an":[2,98,169],"effective":[3],"technique":[4],"for":[5,82,163,177],"reducing":[6],"the":[7,13,35,40,87],"number":[8],"of":[9,15,42,89,148,183],"computations":[10,81],"and":[11,38,122,124,135,141,150,185,206],"improving":[12],"performance":[14,176,188,204],"deep":[16],"neural":[17],"networks":[18],"(DNNs).":[19],"Weight":[20],"quantization":[21,32,53,101,113,131,175,198],"popular":[23],"because":[24],"weights":[25,37,136],"can":[26,46],"be":[27],"trained":[28],"beforehand.":[29],"However,":[30],"weight":[31,151],"only":[33,162],"targets":[34],"kernel":[36],"ignores":[39],"sensitivity":[41,75,139],"input":[43,52,74,149],"features,":[44],"which":[45],"lead":[47],"to":[48,60,96,116,154,172],"reduced":[49],"accuracy.":[50,66],"Fine-grained":[51],"has":[54],"gained":[55],"attention":[56],"as":[57],"a":[58,110,181,195,202,207],"way":[59],"speed":[61],"up":[62],"DNNs":[63],"while":[64],"maintaining":[65],"Existing":[67],"approaches":[68],"determine":[69],"computation":[70],"precision":[71,88],"based":[72],"on":[73],"but":[76],"do":[77],"not":[78],"effectively":[79],"reduce":[80,119],"insensitive":[83],"outputs":[84],"or":[85],"retain":[86],"sensitive":[90,165],"outputs.":[91,166],"These":[92],"limitations":[93],"motivate":[94],"us":[95],"develop":[97],"output-directed":[99],"dynamic":[100],"method":[102],"named":[103],"ODQ":[104,108,174,184,200],"in":[105],"this":[106],"paper.":[107],"two-stage":[111],"DNN":[112],"scheme":[114],"designed":[115,168],"improve":[117,126],"performance,":[118],"energy":[120,209],"consumption,":[121],"maintain":[123],"often":[125],"accuracy,":[127],"compared":[128],"with":[129,194,211],"existing":[130],"methods.":[132],"Specifically,":[133],"inputs":[134],"go":[137],"through":[138],"prediction":[140],"result":[142],"generation.":[143],"The":[144],"high-order":[145],"2":[146],"bits":[147],"are":[152],"used":[153],"predict":[155],"output":[156],"sensitivity.":[157],"Result":[158],"generation":[159],"performed":[161],"predicted":[164],"We":[167,179],"FPGA":[170],"accelerator":[171],"optimize":[173],"DNNs.":[178,192],"implement":[180],"prototype":[182],"evaluate":[186],"its":[187],"using":[189],"several":[190],"state-of-the-art":[191,196],"Compared":[193],"input-directed":[197],"approach,":[199],"achieves":[201],"67.6%":[203],"speedup":[205],"66.9%":[208],"saving,":[210],"minimal":[212],"accuracy":[213],"degradation":[214],"(\u2264":[215],"0.6%).":[216]},"counts_by_year":[{"year":2024,"cited_by_count":2}],"updated_date":"2026-07-31T08:31:51.225901","created_date":"2025-10-10T00:00:00"}