{"id":"https://openalex.org/W4387609117","doi":"https://doi.org/10.1109/tai.2023.3324315","title":"Energy Efficient and Fast CNN Inference by Exploring Weight Approximation and Computational Reuse","display_name":"Energy Efficient and Fast CNN Inference by Exploring Weight Approximation and Computational Reuse","publication_year":2023,"publication_date":"2023-10-13","ids":{"openalex":"https://openalex.org/W4387609117","doi":"https://doi.org/10.1109/tai.2023.3324315"},"language":"en","primary_location":{"id":"doi:10.1109/tai.2023.3324315","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tai.2023.3324315","pdf_url":null,"source":{"id":"https://openalex.org/S4210169448","display_name":"IEEE Transactions on Artificial Intelligence","issn_l":"2691-4581","issn":["2691-4581"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Artificial Intelligence","raw_type":"journal-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/A5091151249","display_name":"Mohammed F. Tolba","orcid":"https://orcid.org/0000-0002-6412-290X"},"institutions":[{"id":"https://openalex.org/I176601375","display_name":"Khalifa University of Science and Technology","ror":"https://ror.org/05hffr360","country_code":"AE","type":"education","lineage":["https://openalex.org/I176601375"]}],"countries":["AE"],"is_corresponding":false,"raw_author_name":"Mohammed F. Tolba","raw_affiliation_strings":["SoC Center, Khalifa University, Abu Dhabi, UAE","SoC Center, Khalifa University, Abu Dhabi, Abu Dhabi, UAE"],"raw_orcid":"https://orcid.org/0000-0002-6412-290X","affiliations":[{"raw_affiliation_string":"SoC Center, Khalifa University, Abu Dhabi, UAE","institution_ids":["https://openalex.org/I176601375"]},{"raw_affiliation_string":"SoC Center, Khalifa University, Abu Dhabi, Abu Dhabi, UAE","institution_ids":["https://openalex.org/I176601375"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5078108496","display_name":"Hani Saleh","orcid":"https://orcid.org/0000-0002-7185-0278"},"institutions":[{"id":"https://openalex.org/I176601375","display_name":"Khalifa University of Science and Technology","ror":"https://ror.org/05hffr360","country_code":"AE","type":"education","lineage":["https://openalex.org/I176601375"]}],"countries":["AE"],"is_corresponding":false,"raw_author_name":"Hani Saleh","raw_affiliation_strings":["SoC Center, Khalifa University, Abu Dhabi, UAE","SoC Center, Khalifa University, Abu Dhabi, Abu Dhabi, UAE"],"raw_orcid":"https://orcid.org/0000-0002-7185-0278","affiliations":[{"raw_affiliation_string":"SoC Center, Khalifa University, Abu Dhabi, UAE","institution_ids":["https://openalex.org/I176601375"]},{"raw_affiliation_string":"SoC Center, Khalifa University, Abu Dhabi, Abu Dhabi, UAE","institution_ids":["https://openalex.org/I176601375"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5013814572","display_name":"Baker Mohammad","orcid":"https://orcid.org/0000-0002-6063-473X"},"institutions":[{"id":"https://openalex.org/I176601375","display_name":"Khalifa University of Science and Technology","ror":"https://ror.org/05hffr360","country_code":"AE","type":"education","lineage":["https://openalex.org/I176601375"]}],"countries":["AE"],"is_corresponding":false,"raw_author_name":"Baker Mohammad","raw_affiliation_strings":["SoC Center, Khalifa University, Abu Dhabi, UAE","SoC Center, Khalifa University, Abu Dhabi, Abu Dhabi, UAE"],"raw_orcid":"https://orcid.org/0000-0002-6063-473X","affiliations":[{"raw_affiliation_string":"SoC Center, Khalifa University, Abu Dhabi, UAE","institution_ids":["https://openalex.org/I176601375"]},{"raw_affiliation_string":"SoC Center, Khalifa University, Abu Dhabi, Abu Dhabi, UAE","institution_ids":["https://openalex.org/I176601375"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5020682177","display_name":"Mahmoud Al\u2010Qutayri","orcid":"https://orcid.org/0000-0002-9600-8036"},"institutions":[{"id":"https://openalex.org/I176601375","display_name":"Khalifa University of Science and Technology","ror":"https://ror.org/05hffr360","country_code":"AE","type":"education","lineage":["https://openalex.org/I176601375"]}],"countries":["AE"],"is_corresponding":false,"raw_author_name":"Mahmoud Al-Qutayri","raw_affiliation_strings":["SoC Center, Khalifa University, Abu Dhabi, UAE","SoC Center, Khalifa University, Abu Dhabi, Abu Dhabi, UAE"],"raw_orcid":"https://orcid.org/0000-0002-9600-8036","affiliations":[{"raw_affiliation_string":"SoC Center, Khalifa University, Abu Dhabi, UAE","institution_ids":["https://openalex.org/I176601375"]},{"raw_affiliation_string":"SoC Center, Khalifa University, Abu Dhabi, Abu Dhabi, UAE","institution_ids":["https://openalex.org/I176601375"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5022750836","display_name":"Ayman Hroub","orcid":"https://orcid.org/0009-0004-8023-6052"},"institutions":[{"id":"https://openalex.org/I94800806","display_name":"Birzeit University","ror":"https://ror.org/0256kw398","country_code":"PS","type":"education","lineage":["https://openalex.org/I94800806"]}],"countries":["PS"],"is_corresponding":false,"raw_author_name":"Ayman Hroub","raw_affiliation_strings":["Department of Electrical and Computer Engineering, Birzeit University, Ramallah, Palestine","Department of Electrical and Computer Engineering Birzeit University, Birzeit, Ramallah, Palestine"],"raw_orcid":"https://orcid.org/0009-0004-8023-6052","affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, Birzeit University, Ramallah, Palestine","institution_ids":["https://openalex.org/I94800806"]},{"raw_affiliation_string":"Department of Electrical and Computer Engineering Birzeit University, Birzeit, Ramallah, Palestine","institution_ids":["https://openalex.org/I94800806"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5045307151","display_name":"T. Stouraitis","orcid":"https://orcid.org/0000-0002-3696-4958"},"institutions":[{"id":"https://openalex.org/I176601375","display_name":"Khalifa University of Science and Technology","ror":"https://ror.org/05hffr360","country_code":"AE","type":"education","lineage":["https://openalex.org/I176601375"]}],"countries":["AE"],"is_corresponding":false,"raw_author_name":"Thanos Stouraitis","raw_affiliation_strings":["SoC Center, Khalifa University, Abu Dhabi, UAE","SoC Center, Khalifa University, Abu Dhabi, Abu Dhabi, UAE"],"raw_orcid":"https://orcid.org/0000-0002-3696-4958","affiliations":[{"raw_affiliation_string":"SoC Center, Khalifa University, Abu Dhabi, UAE","institution_ids":["https://openalex.org/I176601375"]},{"raw_affiliation_string":"SoC Center, Khalifa University, Abu Dhabi, Abu Dhabi, UAE","institution_ids":["https://openalex.org/I176601375"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.3191,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.57582406,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":96},"biblio":{"volume":"5","issue":"6","first_page":"2678","last_page":"2692"},"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/T10502","display_name":"Advanced Memory and Neural Computing","score":0.9990000128746033,"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/T11992","display_name":"CCD and CMOS Imaging Sensors","score":0.9948999881744385,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8173277378082275},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.7076525688171387},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.69586181640625},{"id":"https://openalex.org/keywords/dataflow","display_name":"Dataflow","score":0.6563126444816589},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.632448673248291},{"id":"https://openalex.org/keywords/design-space-exploration","display_name":"Design space exploration","score":0.5793495178222656},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.5382463932037354},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.5144184231758118},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.4961319863796234},{"id":"https://openalex.org/keywords/reduction","display_name":"Reduction (mathematics)","score":0.48331689834594727},{"id":"https://openalex.org/keywords/hardware-acceleration","display_name":"Hardware acceleration","score":0.46475568413734436},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.4575277268886566},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.45015284419059753},{"id":"https://openalex.org/keywords/edge-device","display_name":"Edge device","score":0.4490533471107483},{"id":"https://openalex.org/keywords/high-level-synthesis","display_name":"High-level synthesis","score":0.4169531464576721},{"id":"https://openalex.org/keywords/reuse","display_name":"Reuse","score":0.4133906662464142},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.3221674859523773},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.30416882038116455},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.28908491134643555}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8173277378082275},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.7076525688171387},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.69586181640625},{"id":"https://openalex.org/C96324660","wikidata":"https://www.wikidata.org/wiki/Q205446","display_name":"Dataflow","level":2,"score":0.6563126444816589},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.632448673248291},{"id":"https://openalex.org/C2776221188","wikidata":"https://www.wikidata.org/wiki/Q21072556","display_name":"Design space exploration","level":2,"score":0.5793495178222656},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.5382463932037354},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.5144184231758118},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4961319863796234},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.48331689834594727},{"id":"https://openalex.org/C13164978","wikidata":"https://www.wikidata.org/wiki/Q600158","display_name":"Hardware acceleration","level":3,"score":0.46475568413734436},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4575277268886566},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.45015284419059753},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.4490533471107483},{"id":"https://openalex.org/C58013763","wikidata":"https://www.wikidata.org/wiki/Q5754574","display_name":"High-level synthesis","level":3,"score":0.4169531464576721},{"id":"https://openalex.org/C206588197","wikidata":"https://www.wikidata.org/wiki/Q846574","display_name":"Reuse","level":2,"score":0.4133906662464142},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3221674859523773},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.30416882038116455},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.28908491134643555},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0},{"id":"https://openalex.org/C18903297","wikidata":"https://www.wikidata.org/wiki/Q7150","display_name":"Ecology","level":1,"score":0.0},{"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/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"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}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tai.2023.3324315","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tai.2023.3324315","pdf_url":null,"source":{"id":"https://openalex.org/S4210169448","display_name":"IEEE Transactions on Artificial Intelligence","issn_l":"2691-4581","issn":["2691-4581"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Artificial Intelligence","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Affordable and clean energy","score":0.8899999856948853,"id":"https://metadata.un.org/sdg/7"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":44,"referenced_works":["https://openalex.org/W1999085092","https://openalex.org/W2119144962","https://openalex.org/W2147657366","https://openalex.org/W2147768505","https://openalex.org/W2194775991","https://openalex.org/W2289252105","https://openalex.org/W2601564443","https://openalex.org/W2604319603","https://openalex.org/W2612076670","https://openalex.org/W2618530766","https://openalex.org/W2748818695","https://openalex.org/W2793847918","https://openalex.org/W2798170643","https://openalex.org/W2896409484","https://openalex.org/W2906663849","https://openalex.org/W2945146780","https://openalex.org/W2963594949","https://openalex.org/W3089673885","https://openalex.org/W3104263540","https://openalex.org/W3129147932","https://openalex.org/W3132745255","https://openalex.org/W3134750800","https://openalex.org/W3139203094","https://openalex.org/W3161570989","https://openalex.org/W3161926058","https://openalex.org/W3205311052","https://openalex.org/W3205807913","https://openalex.org/W3211387996","https://openalex.org/W3213510604","https://openalex.org/W4205365435","https://openalex.org/W4206117103","https://openalex.org/W4206533878","https://openalex.org/W4244802472","https://openalex.org/W4285241469","https://openalex.org/W4293193211","https://openalex.org/W4298857966","https://openalex.org/W4312684832","https://openalex.org/W4383501651","https://openalex.org/W4385080007","https://openalex.org/W6637967152","https://openalex.org/W6677580257","https://openalex.org/W6684563725","https://openalex.org/W6725543821","https://openalex.org/W6755843862"],"related_works":["https://openalex.org/W1741246166","https://openalex.org/W1508949720","https://openalex.org/W1855080238","https://openalex.org/W2890557332","https://openalex.org/W4281926497","https://openalex.org/W2274562545","https://openalex.org/W3048824278","https://openalex.org/W2461217932","https://openalex.org/W2077629192","https://openalex.org/W2269990635"],"abstract_inverted_index":{"The":[0,57,119,140],"challenges":[1],"of":[2,53,65,72,129,153,211,238],"Convolutional":[3],"neural":[4],"networks":[5],"(CNNs)":[6],"based":[7],"AI":[8],"inference":[9],"on":[10,145,167],"edge":[11],"devices":[12],"include":[13],"computing":[14],"complexity,":[15],"large":[16],"memory":[17],"requirements,":[18],"and":[19,30,51,83,127,138,191,215,236,242],"high":[20],"power":[21,52,136,235],"consumption.":[22],"Researchers":[23],"have":[24],"pursued":[25],"efficient":[26],"hardware,":[27],"dataflow":[28],"optimization,":[29],"new":[31],"algorithms":[32,142],"to":[33,45,157,203,227],"tackle":[34],"these":[35],"obstacles.":[36],"This":[37,219],"paper":[38],"introduced":[39],"a":[40,146,151,178,197,225],"novel":[41],"acceleration":[42],"methodology":[43],"designed":[44],"significantly":[46],"enhance":[47],"the":[48,61,70,88,102,111,115,125,158,161,193,229],"processing":[49],"speed":[50],"existing":[54],"CNN":[55,67,131,163,240],"architectures.":[56],"proposed":[58,120,141,162,194],"approach":[59,121,195],"leverages":[60],"inherent":[62],"static":[63],"nature":[64],"pre-trained":[66],"weights":[68],"through":[69,124],"prism":[71],"linear":[73],"approximation.":[74],"Different":[75],"weight":[76,90],"approximation":[77],"options":[78],"are":[79,143,165],"evaluated,":[80],"including":[81],"within":[82],"across":[84],"kernels.":[85],"By":[86],"using":[87],"same":[89,103,112],"index":[91,104],"from":[92,105,114],"different":[93,106,230,239],"kernels":[94,107],"for":[95,213,217],"approximation,":[96],"computation":[97],"reuse":[98],"becomes":[99],"possible,":[100],"as":[101,224],"is":[108,122],"sampled":[109],"with":[110,187,209],"data":[113],"input":[116],"feature":[117],"map.":[118],"evaluated":[123],"design":[126],"analysis":[128],"four":[130],"architectures,":[132],"considering":[133],"hardwarere":[134],"sources,":[135],"consumption,":[137],"latency.":[139],"implemented":[144,166],"gem5-based":[147],"RISCV":[148],"simulator,":[149],"demonstrating":[150],"speedup":[152],"approximately":[154],"2\u00d7":[155],"compared":[156,202],"baseline.":[159],"Additionally,":[160],"accelerators":[164],"Xilinx":[168],"Kintex":[169],"7":[170],"Field":[171],"Programmable":[172],"Gate":[173],"Array":[174],"(FPGA),":[175],"resulting":[176],"in":[177,181,200],"50%":[179,198],"reduction":[180,199],"FPGA":[182],"hardware":[183],"resources.":[184],"When":[185],"benchmarked":[186],"AlexNet,":[188],"VGG16,":[189],"SqueezeNet,":[190],"ResNet,":[192],"achieves":[196],"multiplications":[201],"previous":[204],"works":[205],"while":[206],"maintaining":[207],"accuracy":[208],"loss":[210],"0.9%":[212],"VGG16":[214],"3.1%":[216],"AlexNet.":[218],"work":[220],"can":[221],"be":[222],"used":[223],"framework":[226],"evaluate":[228],"tradeoffs":[231],"between":[232],"accuracy,":[233],"latency,":[234],"cost":[237],"architectures":[241],"applications":[243]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":1}],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2025-10-10T00:00:00"}
