{"id":"https://openalex.org/W4387250519","doi":"https://doi.org/10.1109/asap57973.2023.00024","title":"Efficient 1D Grouped Convolution for PyTorch a Case Study: Fast On-Device Fine-Tuning for SqueezeBERT","display_name":"Efficient 1D Grouped Convolution for PyTorch a Case Study: Fast On-Device Fine-Tuning for SqueezeBERT","publication_year":2023,"publication_date":"2023-07-01","ids":{"openalex":"https://openalex.org/W4387250519","doi":"https://doi.org/10.1109/asap57973.2023.00024"},"language":"en","primary_location":{"id":"doi:10.1109/asap57973.2023.00024","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/asap57973.2023.00024","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 IEEE 34th International Conference on Application-specific Systems, Architectures and Processors (ASAP)","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/A5056921752","display_name":"Seyyed Hasan Mozafari","orcid":"https://orcid.org/0000-0002-0360-1747"},"institutions":[{"id":"https://openalex.org/I5023651","display_name":"McGill University","ror":"https://ror.org/01pxwe438","country_code":"CA","type":"education","lineage":["https://openalex.org/I5023651"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"S. H. Mozafari","raw_affiliation_strings":["McGill University,Department of Electrical and Computer Engineering,Montréal,QC,Canada"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"McGill University,Department of Electrical and Computer Engineering,Montréal,QC,Canada","institution_ids":["https://openalex.org/I5023651"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5079264223","display_name":"James J. Clark","orcid":"https://orcid.org/0000-0002-4512-6171"},"institutions":[{"id":"https://openalex.org/I5023651","display_name":"McGill University","ror":"https://ror.org/01pxwe438","country_code":"CA","type":"education","lineage":["https://openalex.org/I5023651"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"J. J. Clark","raw_affiliation_strings":["McGill University,Department of Electrical and Computer Engineering,Montréal,QC,Canada"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"McGill University,Department of Electrical and Computer Engineering,Montréal,QC,Canada","institution_ids":["https://openalex.org/I5023651"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5091600002","display_name":"Warren J. Gross","orcid":"https://orcid.org/0000-0002-6226-6037"},"institutions":[{"id":"https://openalex.org/I5023651","display_name":"McGill University","ror":"https://ror.org/01pxwe438","country_code":"CA","type":"education","lineage":["https://openalex.org/I5023651"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"W. J. Gross","raw_affiliation_strings":["McGill University,Department of Electrical and Computer Engineering,Montréal,QC,Canada"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"McGill University,Department of Electrical and Computer Engineering,Montréal,QC,Canada","institution_ids":["https://openalex.org/I5023651"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5034701871","display_name":"Brett H. Meyer","orcid":"https://orcid.org/0000-0002-6650-3298"},"institutions":[{"id":"https://openalex.org/I5023651","display_name":"McGill University","ror":"https://ror.org/01pxwe438","country_code":"CA","type":"education","lineage":["https://openalex.org/I5023651"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"B. H. Meyer","raw_affiliation_strings":["McGill University,Department of Electrical and Computer Engineering,Montréal,QC,Canada"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"McGill University,Department of Electrical and Computer Engineering,Montréal,QC,Canada","institution_ids":["https://openalex.org/I5023651"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I5023651"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.11536908,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"69","last_page":"75"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9991999864578247,"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":0.9991999864578247,"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/T12808","display_name":"Ferroelectric and Negative Capacitance Devices","score":0.9984999895095825,"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/T10028","display_name":"Topic Modeling","score":0.9968000054359436,"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/computer-science","display_name":"Computer science","score":0.7868976593017578},{"id":"https://openalex.org/keywords/convolution","display_name":"Convolution (computer science)","score":0.7800116539001465},{"id":"https://openalex.org/keywords/locality","display_name":"Locality","score":0.7572087049484253},{"id":"https://openalex.org/keywords/edge-device","display_name":"Edge device","score":0.7255439758300781},{"id":"https://openalex.org/keywords/enhanced-data-rates-for-gsm-evolution","display_name":"Enhanced Data Rates for GSM Evolution","score":0.6838183403015137},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6795650720596313},{"id":"https://openalex.org/keywords/memory-management","display_name":"Memory management","score":0.6264576315879822},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.4997999668121338},{"id":"https://openalex.org/keywords/reduction","display_name":"Reduction (mathematics)","score":0.4840616285800934},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.4509066641330719},{"id":"https://openalex.org/keywords/edge-computing","display_name":"Edge computing","score":0.44848570227622986},{"id":"https://openalex.org/keywords/base","display_name":"Base (topology)","score":0.41550159454345703},{"id":"https://openalex.org/keywords/computational-science","display_name":"Computational science","score":0.4143623113632202},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.4055211544036865},{"id":"https://openalex.org/keywords/theoretical-computer-science","display_name":"Theoretical computer science","score":0.33680421113967896},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3000575304031372},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.2937982678413391},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.12072741985321045},{"id":"https://openalex.org/keywords/semiconductor-memory","display_name":"Semiconductor memory","score":0.11717554926872253},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.07981696724891663}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7868976593017578},{"id":"https://openalex.org/C45347329","wikidata":"https://www.wikidata.org/wiki/Q5166604","display_name":"Convolution (computer science)","level":3,"score":0.7800116539001465},{"id":"https://openalex.org/C2779808786","wikidata":"https://www.wikidata.org/wiki/Q6664603","display_name":"Locality","level":2,"score":0.7572087049484253},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.7255439758300781},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.6838183403015137},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6795650720596313},{"id":"https://openalex.org/C176649486","wikidata":"https://www.wikidata.org/wiki/Q2308807","display_name":"Memory management","level":3,"score":0.6264576315879822},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4997999668121338},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.4840616285800934},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4509066641330719},{"id":"https://openalex.org/C2778456923","wikidata":"https://www.wikidata.org/wiki/Q5337692","display_name":"Edge computing","level":3,"score":0.44848570227622986},{"id":"https://openalex.org/C42058472","wikidata":"https://www.wikidata.org/wiki/Q810214","display_name":"Base (topology)","level":2,"score":0.41550159454345703},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.4143623113632202},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.4055211544036865},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.33680421113967896},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3000575304031372},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.2937982678413391},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.12072741985321045},{"id":"https://openalex.org/C98986596","wikidata":"https://www.wikidata.org/wiki/Q1143031","display_name":"Semiconductor memory","level":2,"score":0.11717554926872253},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.07981696724891663},{"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/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.0},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","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}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/asap57973.2023.00024","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/asap57973.2023.00024","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 IEEE 34th International Conference on Application-specific Systems, Architectures and Processors (ASAP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.5899999737739563,"display_name":"No poverty","id":"https://metadata.un.org/sdg/1"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":33,"referenced_works":["https://openalex.org/W1976188373","https://openalex.org/W2172654076","https://openalex.org/W2291160084","https://openalex.org/W2670151302","https://openalex.org/W2737740651","https://openalex.org/W2896457183","https://openalex.org/W2903697572","https://openalex.org/W2922395136","https://openalex.org/W2923014074","https://openalex.org/W2955425717","https://openalex.org/W2962697884","https://openalex.org/W2962821792","https://openalex.org/W2963125010","https://openalex.org/W2963163009","https://openalex.org/W2996428491","https://openalex.org/W2998506323","https://openalex.org/W3007007518","https://openalex.org/W3034457371","https://openalex.org/W3038012435","https://openalex.org/W3043802286","https://openalex.org/W3046503725","https://openalex.org/W3105645800","https://openalex.org/W3105966348","https://openalex.org/W3170233084","https://openalex.org/W4205256460","https://openalex.org/W4289300273","https://openalex.org/W4295312788","https://openalex.org/W4297775537","https://openalex.org/W4385245566","https://openalex.org/W6737664043","https://openalex.org/W6750615492","https://openalex.org/W6762718338","https://openalex.org/W6766978945"],"related_works":["https://openalex.org/W4322761281","https://openalex.org/W4238233472","https://openalex.org/W3111395152","https://openalex.org/W4313526662","https://openalex.org/W4313463218","https://openalex.org/W4312996489","https://openalex.org/W3106131444","https://openalex.org/W3216099748","https://openalex.org/W4205963435","https://openalex.org/W3214037210"],"abstract_inverted_index":{"Grouped":[0],"convolution":[1,11],"has":[2,130],"been":[3],"observed":[4],"to":[5,61],"be":[6],"an":[7],"effective":[8],"approximation":[9],"for":[10,37,71,122,150],"in":[12,53,93,147],"many":[13],"DNN":[14],"applications.":[15],"For":[16],"example,":[17],"SqueezeBERT,":[18],"which":[19],"is":[20,35,120],"a":[21,131,143],"light":[22],"and":[23,78,108],"fast":[24],"BERT":[25,64],"language":[26],"processing":[27],"model,":[28,65],"utilizes":[29],"1D":[30,90,97,112],"grouped":[31,91,98,113],"convolutions.":[32,114],"Though":[33],"SqueezeBERT":[34,151],"well-optimized":[36],"inference":[38,135],"on":[39,57,134,152],"edge":[40,72,123,153],"devices,":[41],"it":[42,129],"suffers":[43],"from":[44,86],"poor":[45,82,105],"memory":[46,83,106],"management":[47,84],"during":[48],"fine-tuning":[49,55,148],"(training).":[50],"This":[51],"results":[52],"longer":[54],"time":[56,149],"resource-limited":[58],"GPUs":[59],"compared":[60],"the":[62,87,104],"original":[63],"BERT-base,":[66],"despite":[67],"being":[68],"specifically":[69],"designed":[70],"devices.":[73,154],"We":[74,95,115],"study":[75],"this":[76,81],"behavior":[77],"show":[79,116],"that":[80,117],"originates":[85],"use":[88],"of":[89,111],"convolutions":[92,99],"SqueezeBERT.":[94],"re-implement":[96],"using":[100],"fully-connected":[101],"layers,":[102],"addressing":[103],"allocation":[107],"data":[109],"locality":[110],"our":[118,139],"method":[119],"well-suited":[121],"devices":[124],"with":[125],"limited":[126],"memory;":[127],"further,":[128],"negligible":[132],"effect":[133],"speed.":[136],"When":[137],"utilizing":[138],"method,":[140],"we":[141],"observe":[142],"42":[144],"%":[145],"reduction":[146]},"counts_by_year":[],"updated_date":"2026-06-26T08:34:08.712188","created_date":"2025-10-10T00:00:00"}