{"id":"https://openalex.org/W4401635337","doi":"https://doi.org/10.1109/tcad.2024.3445263","title":"Efficient Batched Inference in Conditional Neural Networks","display_name":"Efficient Batched Inference in Conditional Neural Networks","publication_year":2024,"publication_date":"2024-08-16","ids":{"openalex":"https://openalex.org/W4401635337","doi":"https://doi.org/10.1109/tcad.2024.3445263"},"language":"en","primary_location":{"id":"doi:10.1109/tcad.2024.3445263","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcad.2024.3445263","pdf_url":null,"source":{"id":"https://openalex.org/S100835903","display_name":"IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems","issn_l":"0278-0070","issn":["0278-0070","1937-4151"],"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 Computer-Aided Design of Integrated Circuits and Systems","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/A5022503279","display_name":"Surya Selvam","orcid":"https://orcid.org/0000-0002-0300-5478"},"institutions":[{"id":"https://openalex.org/I219193219","display_name":"Purdue University West Lafayette","ror":"https://ror.org/02dqehb95","country_code":"US","type":"education","lineage":["https://openalex.org/I219193219"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Surya Selvam","raw_affiliation_strings":["Elmore Family School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN, USA"],"raw_orcid":"https://orcid.org/0000-0002-0300-5478","affiliations":[{"raw_affiliation_string":"Elmore Family School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN, USA","institution_ids":["https://openalex.org/I219193219"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063948719","display_name":"Amrit Nagarajan","orcid":"https://orcid.org/0000-0002-2847-4721"},"institutions":[{"id":"https://openalex.org/I219193219","display_name":"Purdue University West Lafayette","ror":"https://ror.org/02dqehb95","country_code":"US","type":"education","lineage":["https://openalex.org/I219193219"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Amrit Nagarajan","raw_affiliation_strings":["Elmore Family School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN, USA"],"raw_orcid":"https://orcid.org/0000-0002-2847-4721","affiliations":[{"raw_affiliation_string":"Elmore Family School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN, USA","institution_ids":["https://openalex.org/I219193219"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5065766721","display_name":"Anand Raghunathan","orcid":"https://orcid.org/0000-0002-4624-564X"},"institutions":[{"id":"https://openalex.org/I219193219","display_name":"Purdue University West Lafayette","ror":"https://ror.org/02dqehb95","country_code":"US","type":"education","lineage":["https://openalex.org/I219193219"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Anand Raghunathan","raw_affiliation_strings":["Elmore Family School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Elmore Family School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN, USA","institution_ids":["https://openalex.org/I219193219"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5022503279"],"corresponding_institution_ids":["https://openalex.org/I219193219"],"apc_list":null,"apc_paid":null,"fwci":0.316,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.64091368,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":98},"biblio":{"volume":"43","issue":"11","first_page":"4081","last_page":"4092"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.9746000170707703,"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"}},"topics":[{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.9746000170707703,"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/T10862","display_name":"AI in cancer detection","score":0.9125999808311462,"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/T10057","display_name":"Face and Expression Recognition","score":0.9032999873161316,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.7453882694244385},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5635187029838562},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5298126935958862},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.47507208585739136},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.399831086397171}],"concepts":[{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7453882694244385},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5635187029838562},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5298126935958862},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.47507208585739136},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.399831086397171}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tcad.2024.3445263","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcad.2024.3445263","pdf_url":null,"source":{"id":"https://openalex.org/S100835903","display_name":"IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems","issn_l":"0278-0070","issn":["0278-0070","1937-4151"],"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 Computer-Aided Design of Integrated Circuits and Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":51,"referenced_works":["https://openalex.org/W2194775991","https://openalex.org/W2550821151","https://openalex.org/W2606722458","https://openalex.org/W2798722023","https://openalex.org/W2884751099","https://openalex.org/W2896457183","https://openalex.org/W2951287298","https://openalex.org/W2962677625","https://openalex.org/W2962944050","https://openalex.org/W2963393494","https://openalex.org/W2979826702","https://openalex.org/W3010833785","https://openalex.org/W3034779821","https://openalex.org/W3035038672","https://openalex.org/W3035251378","https://openalex.org/W3086425440","https://openalex.org/W3094502228","https://openalex.org/W3122890974","https://openalex.org/W3126595748","https://openalex.org/W3130554079","https://openalex.org/W3173563887","https://openalex.org/W3204647170","https://openalex.org/W3207622241","https://openalex.org/W4243928383","https://openalex.org/W4252337780","https://openalex.org/W4286901804","https://openalex.org/W4293718192","https://openalex.org/W4297950276","https://openalex.org/W4322718191","https://openalex.org/W4324292875","https://openalex.org/W4372349390","https://openalex.org/W4377130682","https://openalex.org/W4380874786","https://openalex.org/W4382467688","https://openalex.org/W4385245566","https://openalex.org/W4403618014","https://openalex.org/W6679436768","https://openalex.org/W6684191040","https://openalex.org/W6732520560","https://openalex.org/W6751913510","https://openalex.org/W6755207826","https://openalex.org/W6766978945","https://openalex.org/W6778883912","https://openalex.org/W6779068807","https://openalex.org/W6779313456","https://openalex.org/W6786224425","https://openalex.org/W6840061620","https://openalex.org/W6843363710","https://openalex.org/W6847363464","https://openalex.org/W6850625674","https://openalex.org/W6873314389"],"related_works":["https://openalex.org/W2961085424","https://openalex.org/W4306674287","https://openalex.org/W3046775127","https://openalex.org/W3107602296","https://openalex.org/W4394896187","https://openalex.org/W3170094116","https://openalex.org/W4386462264","https://openalex.org/W4364306694","https://openalex.org/W4312192474","https://openalex.org/W4283697347"],"abstract_inverted_index":{"Conditional":[0,92,188],"neural":[1],"networks":[2,5],"(NNs)":[3],"are":[4,28,123],"in":[6,51,148],"which":[7],"the":[8,14,52,158],"computations":[9],"performed":[10],"vary":[11],"based":[12],"on":[13],"input.":[15],"Many":[16],"NNs":[17,93],"of":[18,96,120,161,175,186],"interest":[19],"(such":[20],"as":[21,47],"autoregressive":[22,197],"transformers":[23],"for":[24,91],"sequence":[25],"generation":[26],"tasks)":[27],"inherently":[29],"conditional":[30],"since":[31],"they":[32],"process":[33],"variable-length":[34,38],"inputs":[35,61,121],"or":[36],"produce":[37],"outputs.":[39],"In":[40],"addition,":[41],"popular":[42],"NN":[43,78],"optimization":[44],"techniques,":[45],"such":[46],"early":[48],"exit,":[49],"result":[50],"computational":[53,101,129,133],"footprint":[54],"varying":[55],"across":[56,60,183],"inputs.":[57],"Computational":[58],"irregularity":[59,139],"presents":[62],"a":[63,67,113,149,184],"challenge":[64],"to":[65,71,117,126,166],"batching,":[66],"technique":[68],"widely":[69],"used":[70],"improve":[72],"hardware":[73],"utilization":[74],"and":[75,105,196],"throughput":[76,160],"during":[77],"inference.":[79],"To":[80],"address":[81],"this":[82],"challenge,":[83],"we":[84],"propose":[85],"BatchCond,":[86],"an":[87],"optimized":[88],"batching":[89,103],"framework":[90],"that":[94,122,155],"consists":[95],"two":[97],"key":[98],"steps:":[99],"1)":[100],"similarity-driven":[102],"(SimBatch)":[104],"2)":[106],"adaptive":[107],"batch":[108],"reorganization":[109],"(ABR).":[110],"SimBatch":[111],"utilizes":[112],"lightweight":[114],"DNN":[115],"predictor":[116],"create":[118],"batches":[119,143],"more":[124],"likely":[125],"share":[127],"similar":[128,146],"patterns,":[130],"thereby":[131],"reducing":[132],"irregularity.":[134],"Further,":[135],"ABR":[136],"addresses":[137],"residual":[138],"by":[140,164],"dynamically":[141],"splitting":[142],"into":[144],"computationally":[145],"sub-batches":[147],"hardware-aware":[150],"manner.":[151],"Our":[152],"experiments":[153],"demonstrate":[154],"BatchCond":[156],"improves":[157],"overall":[159],"batched":[162],"inference":[163],"up":[165],"<inline-formula":[167,176],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[168,177,204],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">":[169,178],"<tex-math":[170,179],"notation=\"LaTeX\">$6.6\\times":[171],"$":[172,181],"</tex-math></inline-formula>":[173],"(mean":[174],"notation=\"LaTeX\">$2.5\\times":[180],"</tex-math></inline-formula>)":[182],"suite":[185],"diverse":[187],"NNs,":[189],"including":[190],"early-exit":[191],"networks,":[192,195],"dynamic":[193],"slimmable":[194],"transformers.":[198],"Code":[199],"is":[200],"available":[201],"at":[202],"<uri":[203],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">https://github.com/surya00060/BatchCond</uri>.":[205]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-05-21T09:19:25.381259","created_date":"2025-10-10T00:00:00"}