{"id":"https://openalex.org/W4378800993","doi":"https://doi.org/10.1145/3583781.3590292","title":"Design Space Exploration of Layer-Wise Mixed-Precision Quantization with Tightly Integrated Edge Inference Units","display_name":"Design Space Exploration of Layer-Wise Mixed-Precision Quantization with Tightly Integrated Edge Inference Units","publication_year":2023,"publication_date":"2023-05-31","ids":{"openalex":"https://openalex.org/W4378800993","doi":"https://doi.org/10.1145/3583781.3590292"},"language":"en","primary_location":{"id":"doi:10.1145/3583781.3590292","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3583781.3590292","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Great Lakes Symposium on VLSI 2023","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/A5100772007","display_name":"Xiaotian Zhao","orcid":"https://orcid.org/0009-0001-3088-8229"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiaotian Zhao","raw_affiliation_strings":["Shanghai Jiao Tong University, Shanghai, China"],"raw_orcid":"https://orcid.org/0009-0001-3088-8229","affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University, Shanghai, China","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100631409","display_name":"Y Gao","orcid":"https://orcid.org/0009-0002-9526-6954"},"institutions":[{"id":"https://openalex.org/I51556381","display_name":"University of Virginia","ror":"https://ror.org/0153tk833","country_code":"US","type":"education","lineage":["https://openalex.org/I51556381"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yimin Gao","raw_affiliation_strings":["University of Virginia, Charlottesville, VA, USA"],"raw_orcid":"https://orcid.org/0009-0002-9526-6954","affiliations":[{"raw_affiliation_string":"University of Virginia, Charlottesville, VA, USA","institution_ids":["https://openalex.org/I51556381"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102008354","display_name":"Vaibhav Verma","orcid":"https://orcid.org/0000-0001-9403-6319"},"institutions":[{"id":"https://openalex.org/I51556381","display_name":"University of Virginia","ror":"https://ror.org/0153tk833","country_code":"US","type":"education","lineage":["https://openalex.org/I51556381"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Vaibhav Verma","raw_affiliation_strings":["University of Virginia, Charlottesville, VA, USA"],"raw_orcid":"https://orcid.org/0000-0001-9403-6319","affiliations":[{"raw_affiliation_string":"University of Virginia, Charlottesville, VA, USA","institution_ids":["https://openalex.org/I51556381"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101481055","display_name":"Ruge Xu","orcid":"https://orcid.org/0009-0002-8632-0883"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ruge Xu","raw_affiliation_strings":["Shanghai Jiao Tong University, Shanghai, China"],"raw_orcid":"https://orcid.org/0009-0002-8632-0883","affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University, Shanghai, China","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5068766978","display_name":"Mircea R. Stan","orcid":"https://orcid.org/0000-0003-0577-9976"},"institutions":[{"id":"https://openalex.org/I51556381","display_name":"University of Virginia","ror":"https://ror.org/0153tk833","country_code":"US","type":"education","lineage":["https://openalex.org/I51556381"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Mircea Stan","raw_affiliation_strings":["University of Virginia, Charlottesville, VA, USA"],"raw_orcid":"https://orcid.org/0000-0003-0577-9976","affiliations":[{"raw_affiliation_string":"University of Virginia, Charlottesville, VA, USA","institution_ids":["https://openalex.org/I51556381"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5061878430","display_name":"Xinfei Guo","orcid":"https://orcid.org/0000-0002-2374-3953"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xinfei Guo","raw_affiliation_strings":["Shanghai Jiao Tong University, Shanghai, China"],"raw_orcid":"https://orcid.org/0000-0002-2374-3953","affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University, Shanghai, China","institution_ids":["https://openalex.org/I183067930"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":4,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"467","last_page":"471"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9986000061035156,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9986000061035156,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9937000274658203,"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/T13114","display_name":"Image Processing Techniques and Applications","score":0.9932000041007996,"subfield":{"id":"https://openalex.org/subfields/2214","display_name":"Media Technology"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.7773807048797607},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7097445726394653},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.6910528540611267},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6844450831413269},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.6049471497535706},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.5467946529388428},{"id":"https://openalex.org/keywords/metric","display_name":"Metric (unit)","score":0.4761476516723633},{"id":"https://openalex.org/keywords/edge-device","display_name":"Edge device","score":0.42768406867980957},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.41871920228004456},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.41641905903816223},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.22998127341270447},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.10736969113349915}],"concepts":[{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.7773807048797607},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7097445726394653},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.6910528540611267},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6844450831413269},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.6049471497535706},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.5467946529388428},{"id":"https://openalex.org/C176217482","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Metric (unit)","level":2,"score":0.4761476516723633},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.42768406867980957},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.41871920228004456},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.41641905903816223},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.22998127341270447},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.10736969113349915},{"id":"https://openalex.org/C79974875","wikidata":"https://www.wikidata.org/wiki/Q483639","display_name":"Cloud computing","level":2,"score":0.0},{"id":"https://openalex.org/C21547014","wikidata":"https://www.wikidata.org/wiki/Q1423657","display_name":"Operations management","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}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3583781.3590292","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3583781.3590292","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Great Lakes Symposium on VLSI 2023","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.4300000071525574,"display_name":"Industry, innovation and infrastructure","id":"https://metadata.un.org/sdg/9"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":19,"referenced_works":["https://openalex.org/W2031493867","https://openalex.org/W2117539524","https://openalex.org/W2194775991","https://openalex.org/W2803549871","https://openalex.org/W2882995207","https://openalex.org/W2884150179","https://openalex.org/W2982041622","https://openalex.org/W2982479999","https://openalex.org/W3034940165","https://openalex.org/W3042701997","https://openalex.org/W3042944763","https://openalex.org/W3096533519","https://openalex.org/W3108833523","https://openalex.org/W3128943006","https://openalex.org/W3163516730","https://openalex.org/W3216111572","https://openalex.org/W4224228206","https://openalex.org/W4287711307","https://openalex.org/W4292828577"],"related_works":["https://openalex.org/W4386245174","https://openalex.org/W4200132709","https://openalex.org/W3204400881","https://openalex.org/W3214410901","https://openalex.org/W3204296682","https://openalex.org/W3183118997","https://openalex.org/W2917767146","https://openalex.org/W2752178021","https://openalex.org/W3176282186","https://openalex.org/W2973622361"],"abstract_inverted_index":{"Layer-wise":[0],"mixed-precision":[1],"quantization":[2,24],"(MPQ)":[3],"has":[4],"become":[5],"prevailing":[6],"for":[7],"edge":[8,87],"inference":[9,96,125],"since":[10],"it":[11],"strikes":[12],"a":[13,50,67,152],"better":[14],"balance":[15],"between":[16],"accuracy":[17,74,126],"and":[18,75,106],"efficiency":[19],"compared":[20,131,150],"to":[21,66,91,132,151],"the":[22,43,94,101,116,133,139],"uniform":[23],"scheme.":[25],"Existing":[26],"MPQ":[27,52,85,135,142],"strategies":[28],"either":[29],"lacked":[30],"hardware":[31,78,129],"awareness":[32],"or":[33],"incurred":[34],"huge":[35],"computation":[36],"costs,":[37],"which":[38],"gated":[39],"their":[40],"deployment":[41],"at":[42],"edge.":[44],"In":[45,137],"this":[46],"work,":[47],"we":[48,89],"propose":[49,90],"novel":[51],"search":[53,118],"algorithm":[54,119],"that":[55,71,115],"obtains":[56],"an":[57],"optimal":[58],"scheme":[59],"by":[60],"\"sampling\"":[61],"layer-wise":[62],"sensitivity":[63],"with":[64,127],"respect":[65],"newly":[68],"proposed":[69,117],"metric":[70],"incorporates":[72],"both":[73],"proxy":[76],"of":[77,100,146],"cost.":[79],"To":[80],"further":[81],"efficiently":[82],"deploy":[83],"post-training":[84],"on":[86],"chips,":[88],"tightly":[92,140],"integrate":[93],"quantized":[95],"units":[97,143],"as":[98],"part":[99],"processor":[102],"pipeline":[103],"through":[104],"micro-architecture":[105],"Instruction":[107],"Set":[108],"Architecture":[109],"(ISA)":[110],"co-design.":[111],"Evaluation":[112],"results":[113],"show":[114],"achieves":[120],"3%":[121],"~":[122,148],"11%":[123],"higher":[124],"similar":[128],"cost":[130],"state-of-the-art":[134],"strategies.":[136],"addition,":[138],"integrated":[141],"achieve":[144],"speedup":[145],"15.13x":[147],"29.65x":[149],"baseline":[153],"RISC-V":[154],"processor.":[155]},"counts_by_year":[{"year":2024,"cited_by_count":4}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}