{"id":"https://openalex.org/W2890770683","doi":"https://doi.org/10.1109/icassp.2018.8461702","title":"An Analytical Method to Determine Minimum Per-Layer Precision of Deep Neural Networks","display_name":"An Analytical Method to Determine Minimum Per-Layer Precision of Deep Neural Networks","publication_year":2018,"publication_date":"2018-04-01","ids":{"openalex":"https://openalex.org/W2890770683","doi":"https://doi.org/10.1109/icassp.2018.8461702","mag":"2890770683"},"language":"en","primary_location":{"id":"doi:10.1109/icassp.2018.8461702","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp.2018.8461702","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","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/A5047303294","display_name":"Charbel Sakr","orcid":"https://orcid.org/0000-0001-5641-0541"},"institutions":[{"id":"https://openalex.org/I157725225","display_name":"University of Illinois Urbana-Champaign","ror":"https://ror.org/047426m28","country_code":"US","type":"education","lineage":["https://openalex.org/I157725225"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Charbel Sakr","raw_affiliation_strings":["Dept. of Electrical and Computer Engineering, University of Illinois at Urbana Champaign"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Dept. of Electrical and Computer Engineering, University of Illinois at Urbana Champaign","institution_ids":["https://openalex.org/I157725225"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5057014407","display_name":"Naresh R. Shanbhag","orcid":"https://orcid.org/0000-0002-4323-9164"},"institutions":[{"id":"https://openalex.org/I157725225","display_name":"University of Illinois Urbana-Champaign","ror":"https://ror.org/047426m28","country_code":"US","type":"education","lineage":["https://openalex.org/I157725225"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Naresh Shanbhag","raw_affiliation_strings":["Dept. of Electrical and Computer Engineering, University of Illinois at Urbana Champaign"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Dept. of Electrical and Computer Engineering, University of Illinois at Urbana Champaign","institution_ids":["https://openalex.org/I157725225"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I157725225"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":30,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1090","last_page":"1094"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9997000098228455,"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":0.9997000098228455,"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":0.9994000196456909,"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/T11512","display_name":"Anomaly Detection Techniques and Applications","score":0.9904000163078308,"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/mnist-database","display_name":"MNIST database","score":0.8876965045928955},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7251909971237183},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.6345342397689819},{"id":"https://openalex.org/keywords/reduction","display_name":"Reduction (mathematics)","score":0.5761798620223999},{"id":"https://openalex.org/keywords/layer","display_name":"Layer (electronics)","score":0.5597203969955444},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5252689123153687},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.5194308757781982},{"id":"https://openalex.org/keywords/binary-number","display_name":"Binary number","score":0.5020656585693359},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4719338119029999},{"id":"https://openalex.org/keywords/point","display_name":"Point (geometry)","score":0.4586496353149414},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.45472800731658936},{"id":"https://openalex.org/keywords/state","display_name":"State (computer science)","score":0.4447329044342041},{"id":"https://openalex.org/keywords/computational-complexity-theory","display_name":"Computational complexity theory","score":0.4426983892917633},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.41611364483833313},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.15865954756736755}],"concepts":[{"id":"https://openalex.org/C190502265","wikidata":"https://www.wikidata.org/wiki/Q17069496","display_name":"MNIST database","level":3,"score":0.8876965045928955},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7251909971237183},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.6345342397689819},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.5761798620223999},{"id":"https://openalex.org/C2779227376","wikidata":"https://www.wikidata.org/wiki/Q6505497","display_name":"Layer (electronics)","level":2,"score":0.5597203969955444},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5252689123153687},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.5194308757781982},{"id":"https://openalex.org/C48372109","wikidata":"https://www.wikidata.org/wiki/Q3913","display_name":"Binary number","level":2,"score":0.5020656585693359},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4719338119029999},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.4586496353149414},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.45472800731658936},{"id":"https://openalex.org/C48103436","wikidata":"https://www.wikidata.org/wiki/Q599031","display_name":"State (computer science)","level":2,"score":0.4447329044342041},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.4426983892917633},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.41611364483833313},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.15865954756736755},{"id":"https://openalex.org/C178790620","wikidata":"https://www.wikidata.org/wiki/Q11351","display_name":"Organic chemistry","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},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C94375191","wikidata":"https://www.wikidata.org/wiki/Q11205","display_name":"Arithmetic","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp.2018.8461702","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp.2018.8461702","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.5699999928474426,"id":"https://metadata.un.org/sdg/8","display_name":"Decent work and economic growth"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":23,"referenced_works":["https://openalex.org/W114517082","https://openalex.org/W1522301498","https://openalex.org/W1841592590","https://openalex.org/W2104636679","https://openalex.org/W2145287260","https://openalex.org/W2286365479","https://openalex.org/W2289252105","https://openalex.org/W2300242332","https://openalex.org/W2331143823","https://openalex.org/W2469490737","https://openalex.org/W2741318576","https://openalex.org/W2963674932","https://openalex.org/W2963982496","https://openalex.org/W3118608800","https://openalex.org/W4295262505","https://openalex.org/W6638783484","https://openalex.org/W6675722231","https://openalex.org/W6693397755","https://openalex.org/W6696004547","https://openalex.org/W6698200048","https://openalex.org/W6701947533","https://openalex.org/W6718212895","https://openalex.org/W6742116046"],"related_works":["https://openalex.org/W4385524141","https://openalex.org/W4297776111","https://openalex.org/W3018979822","https://openalex.org/W3026616975","https://openalex.org/W2989784533","https://openalex.org/W4288018014","https://openalex.org/W2996058201","https://openalex.org/W3127679336","https://openalex.org/W4379744580","https://openalex.org/W3042419602"],"abstract_inverted_index":{"There":[0],"has":[1],"been":[2],"growing":[3],"interest":[4],"in":[5,59,73,85,119,169],"the":[6,100,116,120,134,157],"deployment":[7],"of":[8,61,110,137],"deep":[9,53],"learning":[10],"systems":[11],"onto":[12],"resource-constrained":[13],"platforms":[14],"for":[15,91,106],"fast":[16],"and":[17,31,79,102],"efficient":[18],"inference.":[19],"However,":[20],"typical":[21],"models":[22],"are":[23,66,76,160],"overwhelmingly":[24],"complex,":[25],"making":[26],"such":[27,35],"integration":[28],"very":[29],"challenging":[30],"requiring":[32,143],"compression":[33],"mechanisms":[34],"as":[36],"reduced":[37],"precision.":[38],"We":[39],"present":[40],"a":[41,127,138,153],"layer-wise":[42],"granular":[43],"precision":[44,83,118],"analysis":[45],"which":[46],"allows":[47,90],"us":[48],"to":[49,123,145,152,164],"efficiently":[50],"quantize":[51],"pre-trained":[52],"neural":[54],"networks":[55],"at":[56],"minimal":[57],"cost":[58],"terms":[60],"accuracy":[62,135],"degradation.":[63],"Our":[64,88],"results":[65,98],"consistent":[67],"with":[68,166],"recent":[69],"findings":[70],"that":[71],"perturbations":[72],"earlier":[74],"layers":[75],"most":[77],"destructive":[78],"hence":[80],"needing":[81],"more":[82],"than":[84],"later":[86],"layers.":[87],"approach":[89,114],"significant":[92],"complexity":[93,158],"reduction":[94],"demonstrated":[95],"by":[96],"numerical":[97],"on":[99],"MNIST":[101],"CIFAR-10":[103],"datasets.":[104],"Indeed,":[105],"an":[107],"equivalent":[108],"level":[109,136],"accuracy,":[111],"our":[112],"fine-grained":[113],"reduces":[115],"minimum":[117],"network":[121,141],"up":[122,144],"8":[124],"bits":[125],"over":[126],"naive":[128],"uniform":[129],"assignment.":[130],"Furthermore,":[131],"we":[132],"match":[133],"state-of-the-art":[139,154],"binary":[140],"while":[142],"~3.5\u00d7":[146],"lower":[147],"complexity.":[148],"Similarly,":[149],"when":[150],"compared":[151],"fixed-point":[155],"network,":[156],"savings":[159],"even":[161],"higher":[162],"(up":[163],"~14\u00d7)":[165],"no":[167],"loss":[168],"accuracy.":[170]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1},{"year":2023,"cited_by_count":2},{"year":2022,"cited_by_count":3},{"year":2021,"cited_by_count":5},{"year":2020,"cited_by_count":10},{"year":2019,"cited_by_count":6},{"year":2018,"cited_by_count":2}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}