{"id":"https://openalex.org/W2261808795","doi":"https://doi.org/10.1109/hpca.2016.7446050","title":"TABLA: A unified template-based framework for accelerating statistical machine learning","display_name":"TABLA: A unified template-based framework for accelerating statistical machine learning","publication_year":2016,"publication_date":"2016-03-01","ids":{"openalex":"https://openalex.org/W2261808795","doi":"https://doi.org/10.1109/hpca.2016.7446050","mag":"2261808795"},"language":"en","primary_location":{"id":"doi:10.1109/hpca.2016.7446050","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpca.2016.7446050","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2016 IEEE International Symposium on High Performance Computer Architecture (HPCA)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"http://repository.gatech.edu/bitstreams/35a3548f-e6fe-4426-b313-4128c78e00b3/download","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5089590312","display_name":"Divya Mahajan","orcid":"https://orcid.org/0009-0007-8184-0528"},"institutions":[{"id":"https://openalex.org/I130701444","display_name":"Georgia Institute of Technology","ror":"https://ror.org/01zkghx44","country_code":"US","type":"education","lineage":["https://openalex.org/I130701444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Divya Mahajan","raw_affiliation_strings":["Georgia Institute of Technology, Atlanta, GA, US"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Georgia Institute of Technology, Atlanta, GA, US","institution_ids":["https://openalex.org/I130701444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5037553165","display_name":"Jongse Park","orcid":"https://orcid.org/0000-0002-6629-449X"},"institutions":[{"id":"https://openalex.org/I130701444","display_name":"Georgia Institute of Technology","ror":"https://ror.org/01zkghx44","country_code":"US","type":"education","lineage":["https://openalex.org/I130701444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jongse Park","raw_affiliation_strings":["Alternative Computing Technologies (ACT) Lab, Georgia Institute of Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alternative Computing Technologies (ACT) Lab, Georgia Institute of Technology","institution_ids":["https://openalex.org/I130701444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048085877","display_name":"Emmanuel Amaro","orcid":"https://orcid.org/0009-0003-8721-4895"},"institutions":[{"id":"https://openalex.org/I130701444","display_name":"Georgia Institute of Technology","ror":"https://ror.org/01zkghx44","country_code":"US","type":"education","lineage":["https://openalex.org/I130701444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Emmanuel Amaro","raw_affiliation_strings":["Alternative Computing Technologies (ACT) Lab, Georgia Institute of Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alternative Computing Technologies (ACT) Lab, Georgia Institute of Technology","institution_ids":["https://openalex.org/I130701444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5082318034","display_name":"Hardik Sharma","orcid":"https://orcid.org/0000-0003-0028-013X"},"institutions":[{"id":"https://openalex.org/I130701444","display_name":"Georgia Institute of Technology","ror":"https://ror.org/01zkghx44","country_code":"US","type":"education","lineage":["https://openalex.org/I130701444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hardik Sharma","raw_affiliation_strings":["Alternative Computing Technologies (ACT) Lab, Georgia Institute of Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alternative Computing Technologies (ACT) Lab, Georgia Institute of Technology","institution_ids":["https://openalex.org/I130701444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5070172290","display_name":"Amir Yazdanbakhsh","orcid":"https://orcid.org/0000-0001-8199-7671"},"institutions":[{"id":"https://openalex.org/I130701444","display_name":"Georgia Institute of Technology","ror":"https://ror.org/01zkghx44","country_code":"US","type":"education","lineage":["https://openalex.org/I130701444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Amir Yazdanbakhsh","raw_affiliation_strings":["Alternative Computing Technologies (ACT) Lab, Georgia Institute of Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alternative Computing Technologies (ACT) Lab, Georgia Institute of Technology","institution_ids":["https://openalex.org/I130701444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5018716691","display_name":"Joon Kyung Kim","orcid":"https://orcid.org/0000-0002-7984-9669"},"institutions":[{"id":"https://openalex.org/I130701444","display_name":"Georgia Institute of Technology","ror":"https://ror.org/01zkghx44","country_code":"US","type":"education","lineage":["https://openalex.org/I130701444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Joon Kyung Kim","raw_affiliation_strings":["Alternative Computing Technologies (ACT) Lab, Georgia Institute of Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alternative Computing Technologies (ACT) Lab, Georgia Institute of Technology","institution_ids":["https://openalex.org/I130701444"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5084514143","display_name":"Hadi Esmaeilzadeh","orcid":"https://orcid.org/0000-0002-8548-1039"},"institutions":[{"id":"https://openalex.org/I130701444","display_name":"Georgia Institute of Technology","ror":"https://ror.org/01zkghx44","country_code":"US","type":"education","lineage":["https://openalex.org/I130701444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hadi Esmaeilzadeh","raw_affiliation_strings":["Alternative Computing Technologies (ACT) Lab, Georgia Institute of Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alternative Computing Technologies (ACT) Lab, Georgia Institute of Technology","institution_ids":["https://openalex.org/I130701444"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I130701444"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":160,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"14","last_page":"26"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9987000226974487,"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":0.9987000226974487,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9984999895095825,"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/T12072","display_name":"Machine Learning and Algorithms","score":0.9965999722480774,"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":0.8075027465820312},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.6189459562301636},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.6014539003372192},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5475195646286011},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.5452507138252258},{"id":"https://openalex.org/keywords/online-machine-learning","display_name":"Online machine learning","score":0.44131675362586975},{"id":"https://openalex.org/keywords/active-learning","display_name":"Active learning (machine learning)","score":0.41003721952438354},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.39215338230133057},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.34242701530456543},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.32982414960861206},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.2060086727142334}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8075027465820312},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.6189459562301636},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.6014539003372192},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5475195646286011},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.5452507138252258},{"id":"https://openalex.org/C115903097","wikidata":"https://www.wikidata.org/wiki/Q7094097","display_name":"Online machine learning","level":3,"score":0.44131675362586975},{"id":"https://openalex.org/C77967617","wikidata":"https://www.wikidata.org/wiki/Q4677561","display_name":"Active learning (machine learning)","level":2,"score":0.41003721952438354},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.39215338230133057},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.34242701530456543},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.32982414960861206},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.2060086727142334},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/hpca.2016.7446050","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpca.2016.7446050","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2016 IEEE International Symposium on High Performance Computer Architecture (HPCA)","raw_type":"proceedings-article"},{"id":"pmh:oai:smartech.gatech.edu:1853/54043","is_oa":true,"landing_page_url":"http://hdl.handle.net/1853/54043","pdf_url":"http://repository.gatech.edu/bitstreams/35a3548f-e6fe-4426-b313-4128c78e00b3/download","source":{"id":"https://openalex.org/S4377196313","display_name":"SMARTech Repository (Georgia Institute of Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I130701444","host_organization_name":"Georgia Institute of Technology","host_organization_lineage":["https://openalex.org/I130701444"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"Technical Report"}],"best_oa_location":{"id":"pmh:oai:smartech.gatech.edu:1853/54043","is_oa":true,"landing_page_url":"http://hdl.handle.net/1853/54043","pdf_url":"http://repository.gatech.edu/bitstreams/35a3548f-e6fe-4426-b313-4128c78e00b3/download","source":{"id":"https://openalex.org/S4377196313","display_name":"SMARTech Repository (Georgia Institute of Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I130701444","host_organization_name":"Georgia Institute of Technology","host_organization_lineage":["https://openalex.org/I130701444"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"Technical Report"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/9","display_name":"Industry, innovation and infrastructure","score":0.49000000953674316}],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W2261808795.pdf","grobid_xml":"https://content.openalex.org/works/W2261808795.grobid-xml"},"referenced_works_count":68,"referenced_works":["https://openalex.org/W1575115198","https://openalex.org/W1584468971","https://openalex.org/W1667652561","https://openalex.org/W1969102784","https://openalex.org/W1975489482","https://openalex.org/W1982565841","https://openalex.org/W1983157164","https://openalex.org/W1984097780","https://openalex.org/W2000967104","https://openalex.org/W2005487033","https://openalex.org/W2006312753","https://openalex.org/W2019660355","https://openalex.org/W2033882786","https://openalex.org/W2036536766","https://openalex.org/W2043607059","https://openalex.org/W2063459162","https://openalex.org/W2065439108","https://openalex.org/W2066443755","https://openalex.org/W2070192690","https://openalex.org/W2077428505","https://openalex.org/W2090850279","https://openalex.org/W2094286023","https://openalex.org/W2098335003","https://openalex.org/W2099517310","https://openalex.org/W2099705706","https://openalex.org/W2100905558","https://openalex.org/W2105572413","https://openalex.org/W2108199402","https://openalex.org/W2112678088","https://openalex.org/W2112796928","https://openalex.org/W2116267755","https://openalex.org/W2118585731","https://openalex.org/W2119697285","https://openalex.org/W2122522028","https://openalex.org/W2124093496","https://openalex.org/W2125203716","https://openalex.org/W2126760474","https://openalex.org/W2130408605","https://openalex.org/W2132729131","https://openalex.org/W2133156997","https://openalex.org/W2149328392","https://openalex.org/W2149684715","https://openalex.org/W2152839228","https://openalex.org/W2153331583","https://openalex.org/W2153635508","https://openalex.org/W2155893237","https://openalex.org/W2161238936","https://openalex.org/W2163383000","https://openalex.org/W2167008350","https://openalex.org/W2187230075","https://openalex.org/W3003770219","https://openalex.org/W3120421331","https://openalex.org/W3120740533","https://openalex.org/W4236433846","https://openalex.org/W4237280149","https://openalex.org/W4245683599","https://openalex.org/W4250589301","https://openalex.org/W4251597219","https://openalex.org/W4256629673","https://openalex.org/W4301584875","https://openalex.org/W4302296459","https://openalex.org/W6637151318","https://openalex.org/W6651700774","https://openalex.org/W6674977914","https://openalex.org/W6675996895","https://openalex.org/W6677656871","https://openalex.org/W6682335648","https://openalex.org/W6832815480"],"related_works":["https://openalex.org/W1986633584","https://openalex.org/W2118190631","https://openalex.org/W2899532525","https://openalex.org/W4389775782","https://openalex.org/W2037866349","https://openalex.org/W4213079490","https://openalex.org/W2901581617","https://openalex.org/W2913715341","https://openalex.org/W114517082","https://openalex.org/W2165651264"],"abstract_inverted_index":{"A":[0],"growing":[1],"number":[2],"of":[3,47,59,64,87,107,122,197,213,228,237,263,336],"commercial":[4],"and":[5,51,61,76,126,272,275,283,294,301,310,321],"enterprise":[6],"systems":[7],"increasingly":[8],"rely":[9],"on":[10],"compute-intensive":[11,21],"Machine":[12],"Learning":[13],"(ML)":[14],"algorithms.":[15,110,187,199],"While":[16],"the":[17,25,45,57,62,116,139,170,180,205,211,214,225,229,261,299,329],"demand":[18],"for":[19,91,104,135,184,231,246],"these":[20],"applications":[22],"is":[23,113,177],"growing,":[24],"performance":[26],"benefits":[27,262,325],"from":[28],"general-purpose":[29,65],"platforms":[30],"are":[31,326],"diminishing.":[32],"Field":[33],"Programmable":[34],"Gate":[35],"Arrays":[36],"(FPGAs)":[37],"provide":[38,131,292,307],"a":[39,92,99,105,119,132,149,190,201,235,253],"promising":[40],"path":[41],"forward":[42],"to":[43,114,130,193,243,266,316],"accommodate":[44],"needs":[46],"machine":[48,93,108,123],"learning":[49,94,109,124,143,154,186,198,206,249],"algorithms":[50,125,144],"represent":[52],"an":[53,89,157,166],"intermediate":[54],"point":[55],"between":[56],"efficiency":[58],"ASICs":[60],"programmability":[63],"processors.":[66],"However,":[67],"acceleration":[68,265],"with":[69],"FPGAs":[70],"still":[71],"requires":[72],"long":[73],"development":[74],"cycles":[75],"extensive":[77],"expertise":[78],"in":[79,314],"hardware":[80,288],"design.":[81],"To":[82],"tackle":[83],"this":[84,128,195],"challenge,":[85],"instead":[86],"designing":[88],"accelerator":[90,230],"algorithm,":[95],"we":[96],"present":[97],"TABLA,":[98],"framework":[100,192],"that":[101,141,164],"generates":[102,224],"accelerators":[103,245,291,306],"class":[106,196],"The":[111,173],"key":[112],"identify":[115],"commonalities":[117],"across":[118],"wide":[120],"range":[121],"utilize":[127],"commonality":[129],"high-level":[133,219],"abstraction":[134],"programmers.":[136],"TABLA":[137,188],"leverages":[138],"insight":[140],"many":[142],"can":[145,203],"be":[146],"expressed":[147],"as":[148],"stochastic":[150,161],"optimization":[151,158],"problem.":[152],"Therefore,":[153,200],"becomes":[155],"solving":[156],"problem":[159],"using":[160,217,234,286],"gradient":[162,174,212],"descent":[163,175],"minimizes":[165],"objective":[167,181,215],"function":[168,182,216],"over":[169,298],"training":[171],"data.":[172],"solver":[176],"fixed":[178],"while":[179,328],"changes":[183],"different":[185,248],"provides":[189],"template-based":[191],"accelerate":[194],"developer":[202],"specify":[204],"task":[207],"by":[208],"only":[209],"expressing":[210],"our":[218],"language.":[220],"Tabla":[221,242],"then":[222],"automatically":[223],"synthesizable":[226],"implementation":[227],"FPGA":[232,256,264],"realization":[233],"set":[236],"hand-optimized":[238],"templates.":[239],"We":[240,258],"use":[241],"generate":[244],"ten":[247],"tasks":[250],"targeted":[251],"at":[252],"Xilinx":[254],"Zynq":[255],"platform.":[257],"rigorously":[259],"compare":[260],"multi-core":[267],"CPUs":[268],"(ARM":[269],"Cortex":[270],"A15":[271],"Xeon":[273,302],"E3)":[274],"many-core":[276],"GPUs":[277],"(Tegra":[278],"K1,":[279],"GTX":[280,318],"650":[281,319],"Ti,":[282],"Tesla":[284],"K40)":[285],"real":[287],"measurements.":[289],"TABLA-generated":[290],"19.4x":[293],"2.9x":[295],"average":[296],"speedup":[297],"ARM":[300],"processors,":[303],"respectively.":[304,323],"These":[305,324],"17.57x,":[308],"20.2x,":[309],"33.4x":[311],"higher":[312],"Performance-per-Watt":[313],"comparison":[315],"Tegra,":[317],"Ti":[320],"Tesla,":[322],"achieved":[327],"programmers":[330],"write":[331],"less":[332],"than":[333],"50":[334],"lines":[335],"code.":[337]},"counts_by_year":[{"year":2025,"cited_by_count":4},{"year":2024,"cited_by_count":12},{"year":2023,"cited_by_count":13},{"year":2022,"cited_by_count":5},{"year":2021,"cited_by_count":21},{"year":2020,"cited_by_count":30},{"year":2019,"cited_by_count":28},{"year":2018,"cited_by_count":28},{"year":2017,"cited_by_count":13},{"year":2016,"cited_by_count":6}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}