{"id":"https://openalex.org/W4308089797","doi":"https://doi.org/10.1109/secon55815.2022.9918563","title":"Stitching Weight-Shared Deep Neural Networks for Efficient Multitask Inference on GPU","display_name":"Stitching Weight-Shared Deep Neural Networks for Efficient Multitask Inference on GPU","publication_year":2022,"publication_date":"2022-09-20","ids":{"openalex":"https://openalex.org/W4308089797","doi":"https://doi.org/10.1109/secon55815.2022.9918563"},"language":"en","primary_location":{"id":"doi:10.1109/secon55815.2022.9918563","is_oa":false,"landing_page_url":"https://doi.org/10.1109/secon55815.2022.9918563","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2022 19th Annual IEEE International Conference on Sensing, Communication, and Networking (SECON)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://ink.library.smu.edu.sg/sis_research/7486","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100379696","display_name":"Zeyu Wang","orcid":"https://orcid.org/0000-0002-7057-1613"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Zeyu Wang","raw_affiliation_strings":["School of Software and BNRist, Tsinghua University"],"affiliations":[{"raw_affiliation_string":"School of Software and BNRist, Tsinghua University","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102221941","display_name":"Xiaoxi He","orcid":null},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Xiaoxi He","raw_affiliation_strings":["Computer Engineering and Networks Laboratory,ETH Zurich","ETH Zurich, Computer Engineering and Networks Laboratory"],"affiliations":[{"raw_affiliation_string":"Computer Engineering and Networks Laboratory,ETH Zurich","institution_ids":["https://openalex.org/I35440088"]},{"raw_affiliation_string":"ETH Zurich, Computer Engineering and Networks Laboratory","institution_ids":["https://openalex.org/I35440088"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5011140675","display_name":"Zimu Zhou","orcid":"https://orcid.org/0000-0002-5457-6967"},"institutions":[{"id":"https://openalex.org/I79891267","display_name":"Singapore Management University","ror":"https://ror.org/050qmg959","country_code":"SG","type":"education","lineage":["https://openalex.org/I79891267"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Zimu Zhou","raw_affiliation_strings":["School of Computing and Information Systems, Singapore Management University"],"affiliations":[{"raw_affiliation_string":"School of Computing and Information Systems, Singapore Management University","institution_ids":["https://openalex.org/I79891267"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100708433","display_name":"Xu Wang","orcid":"https://orcid.org/0000-0001-7195-5603"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xu Wang","raw_affiliation_strings":["School of Software and BNRist, Tsinghua University"],"affiliations":[{"raw_affiliation_string":"School of Software and BNRist, Tsinghua University","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5080049876","display_name":"Qiang Ma","orcid":"https://orcid.org/0000-0001-7335-4182"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qiang Ma","raw_affiliation_strings":["School of Software and BNRist, Tsinghua University"],"affiliations":[{"raw_affiliation_string":"School of Software and BNRist, Tsinghua University","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100729108","display_name":"Xin Miao","orcid":"https://orcid.org/0000-0002-3208-630X"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xin Miao","raw_affiliation_strings":["School of Software and BNRist, Tsinghua University"],"affiliations":[{"raw_affiliation_string":"School of Software and BNRist, Tsinghua University","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100355531","display_name":"Zhuo Liu","orcid":"https://orcid.org/0000-0002-7591-7775"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhuo Liu","raw_affiliation_strings":["Remarkable Universal Network Company"],"affiliations":[{"raw_affiliation_string":"Remarkable Universal Network Company","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5060999697","display_name":"Lothar Thiele","orcid":"https://orcid.org/0000-0001-6139-868X"},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Lothar Thiele","raw_affiliation_strings":["Computer Engineering and Networks Laboratory,ETH Zurich","ETH Zurich, Computer Engineering and Networks Laboratory"],"affiliations":[{"raw_affiliation_string":"Computer Engineering and Networks Laboratory,ETH Zurich","institution_ids":["https://openalex.org/I35440088"]},{"raw_affiliation_string":"ETH Zurich, Computer Engineering and Networks Laboratory","institution_ids":["https://openalex.org/I35440088"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5089270527","display_name":"Zheng Yang","orcid":"https://orcid.org/0000-0001-8060-7387"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zheng Yang","raw_affiliation_strings":["School of Software and BNRist, Tsinghua University"],"affiliations":[{"raw_affiliation_string":"School of Software and BNRist, Tsinghua University","institution_ids":["https://openalex.org/I99065089"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":9,"corresponding_author_ids":["https://openalex.org/A5100379696"],"corresponding_institution_ids":["https://openalex.org/I99065089"],"apc_list":null,"apc_paid":null,"fwci":0.4084,"has_fulltext":false,"cited_by_count":5,"citation_normalized_percentile":{"value":0.6053901,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":96,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"145","last_page":"153"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9993000030517578,"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":0.9993000030517578,"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/T11273","display_name":"Advanced Graph Neural Networks","score":0.998199999332428,"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/T12702","display_name":"Brain Tumor Detection and Classification","score":0.9916999936103821,"subfield":{"id":"https://openalex.org/subfields/2808","display_name":"Neurology"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8615999817848206},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.7580236196517944},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.6057592034339905},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.5336889028549194},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.5024878978729248},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.4904915988445282},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.38144367933273315},{"id":"https://openalex.org/keywords/theoretical-computer-science","display_name":"Theoretical computer science","score":0.35594040155410767},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.32768869400024414}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8615999817848206},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7580236196517944},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.6057592034339905},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.5336889028549194},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5024878978729248},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.4904915988445282},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.38144367933273315},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.35594040155410767},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.32768869400024414},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/secon55815.2022.9918563","is_oa":false,"landing_page_url":"https://doi.org/10.1109/secon55815.2022.9918563","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2022 19th Annual IEEE International Conference on Sensing, Communication, and Networking (SECON)","raw_type":"proceedings-article"},{"id":"pmh:oai:ink.library.smu.edu.sg:sis_research-8489","is_oa":true,"landing_page_url":"https://ink.library.smu.edu.sg/sis_research/7486","pdf_url":null,"source":{"id":"https://openalex.org/S4377196871","display_name":"Institutional Knowledge (InK) - Institutional Knowledge at Singapore Management University (Singapore Management University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I79891267","host_organization_name":"Singapore Management University","host_organization_lineage":["https://openalex.org/I79891267"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"http://doi.org/10.1109/SECON55815.2022.9918563","raw_type":"Conference Proceeding Article"}],"best_oa_location":{"id":"pmh:oai:ink.library.smu.edu.sg:sis_research-8489","is_oa":true,"landing_page_url":"https://ink.library.smu.edu.sg/sis_research/7486","pdf_url":null,"source":{"id":"https://openalex.org/S4377196871","display_name":"Institutional Knowledge (InK) - Institutional Knowledge at Singapore Management University (Singapore Management University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I79891267","host_organization_name":"Singapore Management University","host_organization_lineage":["https://openalex.org/I79891267"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"http://doi.org/10.1109/SECON55815.2022.9918563","raw_type":"Conference Proceeding Article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/8","display_name":"Decent work and economic growth","score":0.4300000071525574}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W1472213334","https://openalex.org/W1971506073","https://openalex.org/W2139514495","https://openalex.org/W3042419602","https://openalex.org/W2966649771","https://openalex.org/W2078081736","https://openalex.org/W3008625068","https://openalex.org/W2991043486","https://openalex.org/W2993622056","https://openalex.org/W3128807919"],"abstract_inverted_index":{"Intelligent":[0],"personal":[1],"and":[2,189],"home":[3],"applications":[4],"demand":[5],"multiple":[6,25,178],"deep":[7],"neural":[8],"networks":[9],"(DNNs)":[10],"running":[11],"on":[12,53],"resourceconstrained":[13],"platforms":[14],"for":[15,65,74,89,118,137],"compound":[16],"inference":[17,121,169],"tasks,":[18],"known":[19],"as":[20,57],"multitask":[21,120,168],"inference.":[22],"To":[23],"fit":[24],"DNNs":[26,37,139],"into":[27,50],"low-resource":[28],"devices,":[29],"emerging":[30],"techniques":[31],"resort":[32],"to":[33,38,48,152,172,175,185],"weight":[34],"sharing":[35],"among":[36],"reduce":[39],"their":[40],"storage.":[41],"However,":[42],"such":[43,56],"reduction":[44],"in":[45],"storage":[46],"fails":[47],"translate":[49],"efficient":[51,119],"execution":[52,77],"common":[54],"accelerators":[55],"GPUs.":[58],"Most":[59],"DNN":[60],"graph":[61,83,87,116,136,200],"rewriters":[62,84],"are":[63],"blind":[64],"multi-DNN":[66,76,91,199],"optimization,":[67],"while":[68],"GPU":[69,156],"vendors":[70],"provide":[71],"inefficient":[72],"APIs":[73],"parallel":[75],"at":[78],"runtime.":[79],"A":[80],"few":[81],"prior":[82],"suggest":[85],"cross-model":[86],"fusion":[88],"low-latency":[90],"execution.":[92],"Yet":[93],"they":[94],"request":[95],"duplication":[96],"of":[97,105,160],"the":[98,102,155],"shared":[99,143],"weights,":[100],"erasing":[101],"memory":[103,192],"saving":[104],"weight-shared":[106,123,138],"DNNs.":[107,124,161,180],"In":[108],"this":[109],"paper,":[110],"we":[111],"propose":[112],"MTS,":[113],"a":[114,127,133,148,197],"novel":[115],"rewriter":[117],"with":[122,195],"MTS":[125,145,166,181],"adopts":[126],"model":[128,149],"stitching":[129],"algorithm":[130],"which":[131],"outputs":[132],"single":[134],"computational":[135],"without":[140],"duplicating":[141],"any":[142],"weight.":[144],"also":[146,182],"utilizes":[147],"grouping":[150],"strategy":[151],"avoid":[153],"overwhelming":[154],"when":[157],"co-running":[158],"tens":[159],"Extensive":[162],"experiments":[163],"show":[164],"that":[165],"accelerates":[167],"by":[170],"up":[171,184],"6.0\u00d7":[173],"compared":[174,194],"sequentially":[176],"executing":[177],"weightshared":[179],"yields":[183],"2.5\u00d7":[186],"lower":[187],"latency":[188],"3.7\u00d7":[190],"less":[191],"usage":[193],"NETFUSE,":[196],"state-of-the-art":[198],"rewriter.":[201]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2024,"cited_by_count":4}],"updated_date":"2026-04-05T17:49:38.594831","created_date":"2025-10-10T00:00:00"}