{"id":"https://openalex.org/W3119184503","doi":"https://doi.org/10.1117/12.2587045","title":"Almost indirect 8-bit convolution for QNNs","display_name":"Almost indirect 8-bit convolution for QNNs","publication_year":2021,"publication_date":"2021-01-04","ids":{"openalex":"https://openalex.org/W3119184503","doi":"https://doi.org/10.1117/12.2587045","mag":"3119184503"},"language":"en","primary_location":{"id":"doi:10.1117/12.2587045","is_oa":false,"landing_page_url":"https://doi.org/10.1117/12.2587045","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Thirteenth International Conference on Machine Vision","raw_type":"proceedings-article"},"type":"article","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/A5047135094","display_name":"Anton Trusov","orcid":"https://orcid.org/0000-0003-4084-4614"},"institutions":[{"id":"https://openalex.org/I153845743","display_name":"Moscow Institute of Physics and Technology","ror":"https://ror.org/00v0z9322","country_code":"RU","type":"education","lineage":["https://openalex.org/I153845743"]}],"countries":["RU"],"is_corresponding":true,"raw_author_name":"Anton Trusov","raw_affiliation_strings":["Moscow Institute of Physics and Technology (Russian Federation)","Smart Engines Service LLC (Russian Federation)"],"affiliations":[{"raw_affiliation_string":"Moscow Institute of Physics and Technology (Russian Federation)","institution_ids":["https://openalex.org/I153845743"]},{"raw_affiliation_string":"Smart Engines Service LLC (Russian Federation)","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5077963262","display_name":"Elena Limonova","orcid":"https://orcid.org/0000-0001-7673-9109"},"institutions":[{"id":"https://openalex.org/I153845743","display_name":"Moscow Institute of Physics and Technology","ror":"https://ror.org/00v0z9322","country_code":"RU","type":"education","lineage":["https://openalex.org/I153845743"]},{"id":"https://openalex.org/I4210115688","display_name":"Institute for Systems Analysis","ror":"https://ror.org/02kmet255","country_code":"RU","type":"facility","lineage":["https://openalex.org/I4210115688"]}],"countries":["RU"],"is_corresponding":false,"raw_author_name":"Elena Limonova","raw_affiliation_strings":["Institute for Systems Analysis, FRC CSC RAS (Russian Federation)","Moscow Institute of Physics and Technology (Russian Federation)","Smart Engines LLC (Russian Federation)"],"affiliations":[{"raw_affiliation_string":"Institute for Systems Analysis, FRC CSC RAS (Russian Federation)","institution_ids":["https://openalex.org/I4210115688"]},{"raw_affiliation_string":"Moscow Institute of Physics and Technology (Russian Federation)","institution_ids":["https://openalex.org/I153845743"]},{"raw_affiliation_string":"Smart Engines LLC (Russian Federation)","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5006759229","display_name":"Sergey A. Usilin","orcid":"https://orcid.org/0000-0001-6383-9696"},"institutions":[{"id":"https://openalex.org/I4210115688","display_name":"Institute for Systems Analysis","ror":"https://ror.org/02kmet255","country_code":"RU","type":"facility","lineage":["https://openalex.org/I4210115688"]}],"countries":["RU"],"is_corresponding":false,"raw_author_name":"Sergey Usilin","raw_affiliation_strings":["Institute for Systems Analysis FRC CSC RAS (Russian Federation)","Smart Engines Service LLC (Russian Federation)"],"affiliations":[{"raw_affiliation_string":"Institute for Systems Analysis FRC CSC RAS (Russian Federation)","institution_ids":["https://openalex.org/I4210115688"]},{"raw_affiliation_string":"Smart Engines Service LLC (Russian Federation)","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5047135094"],"corresponding_institution_ids":["https://openalex.org/I153845743"],"apc_list":null,"apc_paid":null,"fwci":0.4011,"has_fulltext":false,"cited_by_count":4,"citation_normalized_percentile":{"value":0.58398977,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"39","last_page":"39"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12808","display_name":"Ferroelectric and Negative Capacitance Devices","score":0.9995999932289124,"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"}},"topics":[{"id":"https://openalex.org/T12808","display_name":"Ferroelectric and Negative Capacitance Devices","score":0.9995999932289124,"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/T10320","display_name":"Neural Networks and Applications","score":0.9962000250816345,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9955999851226807,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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":0.7148710489273071},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.5715337991714478},{"id":"https://openalex.org/keywords/kernel","display_name":"Kernel (algebra)","score":0.5333393216133118},{"id":"https://openalex.org/keywords/convolution","display_name":"Convolution (computer science)","score":0.5273345708847046},{"id":"https://openalex.org/keywords/matrix-multiplication","display_name":"Matrix multiplication","score":0.5097483992576599},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.48581552505493164},{"id":"https://openalex.org/keywords/multiplication","display_name":"Multiplication (music)","score":0.47371214628219604},{"id":"https://openalex.org/keywords/memory-footprint","display_name":"Memory footprint","score":0.45146772265434265},{"id":"https://openalex.org/keywords/block","display_name":"Block (permutation group theory)","score":0.42754215002059937},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.23429903388023376},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.13792943954467773},{"id":"https://openalex.org/keywords/discrete-mathematics","display_name":"Discrete mathematics","score":0.10173496603965759},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.08848753571510315}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7148710489273071},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5715337991714478},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.5333393216133118},{"id":"https://openalex.org/C45347329","wikidata":"https://www.wikidata.org/wiki/Q5166604","display_name":"Convolution (computer science)","level":3,"score":0.5273345708847046},{"id":"https://openalex.org/C17349429","wikidata":"https://www.wikidata.org/wiki/Q1049914","display_name":"Matrix multiplication","level":3,"score":0.5097483992576599},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.48581552505493164},{"id":"https://openalex.org/C2780595030","wikidata":"https://www.wikidata.org/wiki/Q3860309","display_name":"Multiplication (music)","level":2,"score":0.47371214628219604},{"id":"https://openalex.org/C74912251","wikidata":"https://www.wikidata.org/wiki/Q6815727","display_name":"Memory footprint","level":2,"score":0.45146772265434265},{"id":"https://openalex.org/C2777210771","wikidata":"https://www.wikidata.org/wiki/Q4927124","display_name":"Block (permutation group theory)","level":2,"score":0.42754215002059937},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.23429903388023376},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.13792943954467773},{"id":"https://openalex.org/C118615104","wikidata":"https://www.wikidata.org/wiki/Q121416","display_name":"Discrete mathematics","level":1,"score":0.10173496603965759},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.08848753571510315},{"id":"https://openalex.org/C84114770","wikidata":"https://www.wikidata.org/wiki/Q46344","display_name":"Quantum","level":2,"score":0.0},{"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/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1117/12.2587045","is_oa":false,"landing_page_url":"https://doi.org/10.1117/12.2587045","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Thirteenth International Conference on Machine Vision","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W3099313426","https://openalex.org/W2017990332","https://openalex.org/W2080337923","https://openalex.org/W4287593139","https://openalex.org/W1488776355","https://openalex.org/W2093251826","https://openalex.org/W752783541","https://openalex.org/W4244262766","https://openalex.org/W1506547947","https://openalex.org/W3034421924"],"abstract_inverted_index":{"The":[0],"implementations":[1],"of":[2,60,73,105,139,147,155,174,192],"the":[3,45,52,71,84,94,116,148],"convolution":[4,164],"operation":[5],"in":[6,70,83,110,137],"neural":[7],"networks":[8,195],"are":[9],"usually":[10],"based":[11],"on":[12,158,196],"convolution-to-GeMM":[13],"(General":[14],"Matrix":[15],"Multiplication)":[16],"transformation.":[17],"However,":[18,65],"this":[19,40,90],"transformation":[20],"requires":[21],"a":[22,56,143,171],"big":[23],"intermediate":[24],"buffer":[25,54,59],"(called":[26],"im2col":[27],"or":[28],"im2row),":[29],"and":[30,36,135,176],"its":[31],"initialization":[32],"is":[33,108,186],"both":[34],"memory":[35,145],"time-consuming.":[37],"To":[38,88],"overcome":[39,89],"problem,":[41,91],"one":[42,122],"may":[43],"use":[44],"Indirect":[46,96,149,169],"Convolution":[47,97,150],"Algorithm.":[48,151],"This":[49,182],"algorithm":[50],"replaces":[51],"im2row":[53,125],"with":[55,142,153],"much":[57],"smaller":[58],"pointers,":[61],"called":[62],"indirection":[63,114],"buffer.":[64,126],"it":[66,185],"limits":[67],"our":[68,163],"flexibility":[69,136],"choice":[72],"multiplication":[74,78],"micro-kernel,":[75],"making":[76],"matrix":[77,111],"slightly":[79],"less":[80],"efficient":[81],"than":[82,168,179],"classical":[85,180],"GeMM":[86,119,140],"algorithm.":[87],"we":[92],"propose":[93],"Almost":[95],"Algorithm,":[98],"which":[99,107],"initializes":[100,121],"small":[101,144,172],"specifically":[102],"ordered":[103],"block":[104,123],"values,":[106],"used":[109],"multiplication,":[112],"via":[113],"buffer,":[115],"same":[117],"way":[118],"Algorithms":[120],"from":[124],"Our":[127],"approach":[128],"allows":[129],"us":[130],"to":[131],"combine":[132],"computational":[133],"efficiency":[134],"shape":[138],"micro-kernels":[141],"footprint":[146],"Experiments":[152],"convolutions":[154],"8-bit":[156,193],"matrices":[157],"ARM":[159],"processors":[160],"show":[161],"that":[162,184],"works":[165],"14-24%":[166],"faster":[167,178],"for":[170,189],"number":[173],"channels":[175],"10-20%":[177],"GeMM-based.":[181],"proves":[183],"perfectly":[187],"suitable":[188],"computing":[190],"inference":[191],"quantized":[194],"mobile":[197],"devices.":[198]},"counts_by_year":[{"year":2023,"cited_by_count":3},{"year":2022,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}