{"id":"https://openalex.org/W4200122815","doi":"https://doi.org/10.1109/iccad51958.2021.9643487","title":"AutoGTCO: Graph and Tensor Co-Optimize for Image Recognition with Transformers on GPU","display_name":"AutoGTCO: Graph and Tensor Co-Optimize for Image Recognition with Transformers on GPU","publication_year":2021,"publication_date":"2021-11-01","ids":{"openalex":"https://openalex.org/W4200122815","doi":"https://doi.org/10.1109/iccad51958.2021.9643487"},"language":"en","primary_location":{"id":"doi:10.1109/iccad51958.2021.9643487","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccad51958.2021.9643487","pdf_url":null,"source":{"id":"https://openalex.org/S4363608354","display_name":"2021 IEEE/ACM International Conference On Computer Aided Design (ICCAD)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 IEEE/ACM International Conference On Computer Aided Design (ICCAD)","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/A5025914107","display_name":"Yang Bai","orcid":"https://orcid.org/0000-0002-3789-0540"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Yang Bai","raw_affiliation_strings":["The Chinese University of Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5091896070","display_name":"Xufeng Yao","orcid":"https://orcid.org/0000-0003-2415-4488"},"institutions":[{"id":"https://openalex.org/I4210127432","display_name":"SMART Reading","ror":"https://ror.org/0345xq683","country_code":"US","type":"nonprofit","lineage":["https://openalex.org/I4210127432"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xufeng Yao","raw_affiliation_strings":["SmartMore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"SmartMore","institution_ids":["https://openalex.org/I4210127432"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100427492","display_name":"Qi Sun","orcid":"https://orcid.org/0000-0001-5153-6698"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Qi Sun","raw_affiliation_strings":["The Chinese University of Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5051340429","display_name":"Bei Yu","orcid":"https://orcid.org/0000-0001-6406-4810"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Bei Yu","raw_affiliation_strings":["The Chinese University of Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong","institution_ids":["https://openalex.org/I177725633"]}]}],"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":7,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"9"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9988999962806702,"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"}},"topics":[{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9988999962806702,"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"}},{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9984999895095825,"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/T12303","display_name":"Tensor decomposition and applications","score":0.9556000232696533,"subfield":{"id":"https://openalex.org/subfields/2605","display_name":"Computational Mathematics"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8004988431930542},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5068376660346985},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.5055930614471436},{"id":"https://openalex.org/keywords/cuda","display_name":"CUDA","score":0.46546459197998047},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.46128883957862854},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.4594602882862091},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.45172280073165894},{"id":"https://openalex.org/keywords/theoretical-computer-science","display_name":"Theoretical computer science","score":0.4498617947101593},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.4471476674079895},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.34505289793014526},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.2451171576976776}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8004988431930542},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5068376660346985},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.5055930614471436},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.46546459197998047},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.46128883957862854},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.4594602882862091},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.45172280073165894},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.4498617947101593},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.4471476674079895},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.34505289793014526},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2451171576976776},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/iccad51958.2021.9643487","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccad51958.2021.9643487","pdf_url":null,"source":{"id":"https://openalex.org/S4363608354","display_name":"2021 IEEE/ACM International Conference On Computer Aided Design (ICCAD)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 IEEE/ACM International Conference On Computer Aided Design (ICCAD)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":48,"referenced_works":["https://openalex.org/W639708223","https://openalex.org/W1486289306","https://openalex.org/W1686810756","https://openalex.org/W1836465849","https://openalex.org/W1903029394","https://openalex.org/W1989988531","https://openalex.org/W2108598243","https://openalex.org/W2155893237","https://openalex.org/W2183341477","https://openalex.org/W2194775991","https://openalex.org/W2295598076","https://openalex.org/W2792643794","https://openalex.org/W2804032941","https://openalex.org/W2804500013","https://openalex.org/W2899771611","https://openalex.org/W2945146780","https://openalex.org/W2953384591","https://openalex.org/W2962856739","https://openalex.org/W2965058486","https://openalex.org/W2981758446","https://openalex.org/W3004659153","https://openalex.org/W3012249773","https://openalex.org/W3091795552","https://openalex.org/W3094502228","https://openalex.org/W3096395190","https://openalex.org/W3096609285","https://openalex.org/W3136585513","https://openalex.org/W3159481202","https://openalex.org/W3166103364","https://openalex.org/W3170841864","https://openalex.org/W3185695382","https://openalex.org/W4206706211","https://openalex.org/W4302296459","https://openalex.org/W4385245566","https://openalex.org/W6620707391","https://openalex.org/W6637373629","https://openalex.org/W6638667902","https://openalex.org/W6713134421","https://openalex.org/W6739901393","https://openalex.org/W6749029207","https://openalex.org/W6751349269","https://openalex.org/W6752057402","https://openalex.org/W6756040250","https://openalex.org/W6778485988","https://openalex.org/W6779728309","https://openalex.org/W6780226713","https://openalex.org/W6785429063","https://openalex.org/W6799167072"],"related_works":["https://openalex.org/W3062287","https://openalex.org/W2380390332","https://openalex.org/W3213381848","https://openalex.org/W2017587301","https://openalex.org/W2012954338","https://openalex.org/W2005148983","https://openalex.org/W2096672917","https://openalex.org/W2392023973","https://openalex.org/W2939411666","https://openalex.org/W4390846322"],"abstract_inverted_index":{"Performance":[0],"optimization":[1,85],"is":[2,45],"the":[3,22,72,84,90,107,121,165],"art":[4],"of":[5,86,106,134],"continuously":[6],"seeking":[7],"an":[8,102],"effective":[9,103],"mapping":[10],"between":[11],"algorithm":[12],"and":[13,114,125,145],"hardware.":[14],"Existing":[15],"deep":[16,51,170],"learning":[17,52,171],"compilers":[18],"or":[19,54],"frameworks":[20],"optimize":[21],"computation":[23,139],"graph":[24],"via":[25],"adapting":[26],"transformations":[27],"manually":[28],"designed":[29],"by":[30,174],"expert":[31],"efforts.":[32],"We":[33],"argue":[34],"that":[35,162],"these":[36],"methods":[37],"ignore":[38],"some":[39],"possible":[40],"graph-level":[41],"optimizations,":[42],"thus":[43],"it":[44,141],"difficult":[46],"to":[47,49,100,169,176],"generalize":[48],"emerging":[50],"models":[53],"new":[55,110],"operators.":[56],"In":[57],"this":[58],"work,":[59],"we":[60],"propose":[61],"AutoGTCO,":[62],"a":[63,94,115],"tensor":[64],"program":[65],"generation":[66,112],"system":[67],"for":[68,120],"vision":[69,159],"tasks":[70,160],"with":[71,78],"transformer":[73,91],"architecture":[74],"on":[75,155],"GPU.":[76],"Compared":[77],"existing":[79],"fusion":[80,88],"strategies,":[81],"AutoGTCO":[82,163],"explores":[83],"operator":[87],"in":[89,128],"model":[92],"through":[93],"novel":[95],"dynamic":[96],"programming":[97],"algorithm.":[98],"Specifically,":[99],"construct":[101],"search":[104,116],"space":[105],"sampled":[108],"programs,":[109],"sketch":[111],"rules":[113],"policy":[117],"are":[118,132],"proposed":[119],"batch":[122],"matrix":[123],"multiplication":[124],"softmax":[126],"operators":[127],"each":[129],"subgraph,":[130],"which":[131],"capable":[133],"fusing":[135],"them":[136,147],"into":[137,148],"large":[138],"units,":[140],"can":[142],"then":[143],"map":[144],"transform":[146],"efficient":[149],"CUDA":[150],"kernels.":[151],"Overall,":[152],"our":[153],"evaluation":[154],"three":[156],"real-world":[157],"transformer-based":[158],"shows":[161],"improves":[164],"execution":[166],"performance":[167],"relative":[168],"engine":[172],"TensorRT":[173],"up":[175],"1.38":[177],"\u00d7.":[178]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2023,"cited_by_count":5}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}