{"id":"https://openalex.org/W4417169303","doi":"https://doi.org/10.1109/dsd67783.2025.00094","title":"DSEParted: Co-Optimization of Embedded NPU Architectures and Neural Network Partitioning","display_name":"DSEParted: Co-Optimization of Embedded NPU Architectures and Neural Network Partitioning","publication_year":2025,"publication_date":"2025-09-10","ids":{"openalex":"https://openalex.org/W4417169303","doi":"https://doi.org/10.1109/dsd67783.2025.00094"},"language":null,"primary_location":{"id":"doi:10.1109/dsd67783.2025.00094","is_oa":false,"landing_page_url":"https://doi.org/10.1109/dsd67783.2025.00094","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 28th Euromicro Conference on Digital System Design (DSD)","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/A5083696598","display_name":"Patrick Schmidt","orcid":"https://orcid.org/0000-0002-8727-6127"},"institutions":[{"id":"https://openalex.org/I102335020","display_name":"Karlsruhe Institute of Technology","ror":"https://ror.org/04t3en479","country_code":"DE","type":"education","lineage":["https://openalex.org/I102335020","https://openalex.org/I1305996414"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Patrick Schmidt","raw_affiliation_strings":["Karlsruhe Institute of Technology,Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Karlsruhe Institute of Technology,Germany","institution_ids":["https://openalex.org/I102335020"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120734492","display_name":"Fabian Krey","orcid":null},"institutions":[{"id":"https://openalex.org/I102335020","display_name":"Karlsruhe Institute of Technology","ror":"https://ror.org/04t3en479","country_code":"DE","type":"education","lineage":["https://openalex.org/I102335020","https://openalex.org/I1305996414"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Fabian Krey","raw_affiliation_strings":["Karlsruhe Institute of Technology,Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Karlsruhe Institute of Technology,Germany","institution_ids":["https://openalex.org/I102335020"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114068443","display_name":"Alexey Serdyuk","orcid":null},"institutions":[{"id":"https://openalex.org/I102335020","display_name":"Karlsruhe Institute of Technology","ror":"https://ror.org/04t3en479","country_code":"DE","type":"education","lineage":["https://openalex.org/I102335020","https://openalex.org/I1305996414"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Alexey Serdyuk","raw_affiliation_strings":["Karlsruhe Institute of Technology,Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Karlsruhe Institute of Technology,Germany","institution_ids":["https://openalex.org/I102335020"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102891258","display_name":"Matthias Stammler","orcid":"https://orcid.org/0009-0006-8843-1076"},"institutions":[{"id":"https://openalex.org/I102335020","display_name":"Karlsruhe Institute of Technology","ror":"https://ror.org/04t3en479","country_code":"DE","type":"education","lineage":["https://openalex.org/I102335020","https://openalex.org/I1305996414"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Matthias Stammler","raw_affiliation_strings":["Karlsruhe Institute of Technology,Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Karlsruhe Institute of Technology,Germany","institution_ids":["https://openalex.org/I102335020"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5053259767","display_name":"Tanja Harbaum","orcid":"https://orcid.org/0000-0001-7310-567X"},"institutions":[{"id":"https://openalex.org/I102335020","display_name":"Karlsruhe Institute of Technology","ror":"https://ror.org/04t3en479","country_code":"DE","type":"education","lineage":["https://openalex.org/I102335020","https://openalex.org/I1305996414"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Tanja Harbaum","raw_affiliation_strings":["Karlsruhe Institute of Technology,Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Karlsruhe Institute of Technology,Germany","institution_ids":["https://openalex.org/I102335020"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5024739574","display_name":"J\u00fcrgen Becker","orcid":"https://orcid.org/0000-0002-5082-5487"},"institutions":[{"id":"https://openalex.org/I102335020","display_name":"Karlsruhe Institute of Technology","ror":"https://ror.org/04t3en479","country_code":"DE","type":"education","lineage":["https://openalex.org/I102335020","https://openalex.org/I1305996414"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"J\u00fcrgen Becker","raw_affiliation_strings":["Karlsruhe Institute of Technology,Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Karlsruhe Institute of Technology,Germany","institution_ids":["https://openalex.org/I102335020"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I102335020"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.45315332,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"653","last_page":"660"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9358999729156494,"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":0.9358999729156494,"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/T14347","display_name":"Big Data and Digital Economy","score":0.01730000041425228,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10904","display_name":"Embedded Systems Design Techniques","score":0.011500000022351742,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.5658000111579895},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5519999861717224},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.5493999719619751},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.47690001130104065},{"id":"https://openalex.org/keywords/efficient-energy-use","display_name":"Efficient energy use","score":0.413100004196167},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.3937000036239624},{"id":"https://openalex.org/keywords/domain","display_name":"Domain (mathematical analysis)","score":0.3659999966621399},{"id":"https://openalex.org/keywords/systems-design","display_name":"Systems design","score":0.3375999927520752}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8167999982833862},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.5658000111579895},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5519999861717224},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.5493999719619751},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.47690001130104065},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.4296000003814697},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.41359999775886536},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.413100004196167},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.3937000036239624},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.37529999017715454},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.3659999966621399},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.34950000047683716},{"id":"https://openalex.org/C31352089","wikidata":"https://www.wikidata.org/wiki/Q3750474","display_name":"Systems design","level":2,"score":0.3375999927520752},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.33149999380111694},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.31709998846054077},{"id":"https://openalex.org/C193415008","wikidata":"https://www.wikidata.org/wiki/Q639681","display_name":"Network architecture","level":2,"score":0.30489999055862427},{"id":"https://openalex.org/C98025372","wikidata":"https://www.wikidata.org/wiki/Q477538","display_name":"Systems architecture","level":3,"score":0.2906999886035919},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.2883000075817108},{"id":"https://openalex.org/C2780165032","wikidata":"https://www.wikidata.org/wiki/Q16869822","display_name":"Energy consumption","level":2,"score":0.2793000042438507},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.2766999900341034},{"id":"https://openalex.org/C2989134064","wikidata":"https://www.wikidata.org/wiki/Q288510","display_name":"Execution time","level":2,"score":0.2752000093460083},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.26589998602867126},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.2565999925136566}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/dsd67783.2025.00094","is_oa":false,"landing_page_url":"https://doi.org/10.1109/dsd67783.2025.00094","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 28th Euromicro Conference on Digital System Design (DSD)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320311687","display_name":"Ministry of Education","ror":"https://ror.org/03m01yf64"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":21,"referenced_works":["https://openalex.org/W2097117768","https://openalex.org/W2126105956","https://openalex.org/W2194775991","https://openalex.org/W2289252105","https://openalex.org/W2549139847","https://openalex.org/W2940862705","https://openalex.org/W2964223234","https://openalex.org/W2980104813","https://openalex.org/W3017521908","https://openalex.org/W3021613070","https://openalex.org/W3102790199","https://openalex.org/W3213528054","https://openalex.org/W4283643682","https://openalex.org/W4293024010","https://openalex.org/W4296209134","https://openalex.org/W4308083753","https://openalex.org/W4362723189","https://openalex.org/W4379115852","https://openalex.org/W4399486726","https://openalex.org/W4401568138","https://openalex.org/W4410887396"],"related_works":[],"abstract_inverted_index":{"Convolutional":[0],"Neural":[1],"Networks":[2],"(CNNs)":[3],"have":[4],"become":[5],"an":[6,70,127,235],"essential":[7],"tool":[8],"in":[9,205],"the":[10,47,64,73,103,121,169,196,199,206,217,221,228,231],"domain":[11],"of":[12,23,36,49,66,105,123,137,198,208,230],"vision":[13],"processing.":[14],"However,":[15,76],"dedicated":[16],"accelerators":[17,38,218],"are":[18],"needed":[19],"for":[20,27,133,151,159,165],"energy-efficient":[21],"execution":[22,65],"these":[24,37,50],"networks,":[25],"especially":[26,85],"embedded":[28],"devices":[29],"with":[30,72,174],"tight":[31],"energy":[32,180],"constraints.":[33],"Integrating":[34],"multiple":[35,59],"via":[39],"chiplets":[40],"promises":[41],"a":[42,55,78,82,114,134,138,148,183,212],"way":[43],"to":[44,118,182,203,211],"scale":[45],"up":[46,202],"performance":[48],"emerging":[51],"systems":[52],"by":[53,155,172,201],"partitioning":[54,77,107],"neural":[56,79,139],"network":[57,80,106],"across":[58],"accelerators.":[60],"This":[61],"approach":[62,117],"enables":[63],"different":[67,87],"layers":[68],"on":[69,177,227],"accelerator":[71,88],"best-suited":[74],"dataflow.":[75],"is":[81,131,241],"non-trivial":[83],"task,":[84],"when":[86,157,163],"architectures":[89],"must":[90],"be":[91],"considered.":[92],"In":[93,161],"this":[94],"paper,":[95],"we":[96,187],"propose":[97],"our":[98,144,190],"framework":[99,145],"DSEParted,":[100],"which":[101],"automates":[102],"co-design":[104],"and":[108],"hardware":[109],"architecture":[110],"optimization.":[111],"It":[112],"employs":[113],"hierarchical":[115],"optimization":[116],"gradually":[119],"reduce":[120,195],"number":[122],"design":[124,147],"candidates":[125],"until":[126],"optimal":[128],"system":[129,149,170,200,233],"configuration":[130],"found":[132],"partitioned":[135],"computation":[136],"network.":[140],"We":[141],"demonstrate":[142],"that":[143,189,214],"can":[146,194],"that,":[150],"GoogLeNet,":[152],"reduces":[153,168],"latency":[154,178],"22.5%":[156],"optimizing":[158,164],"latency.":[160],"addition,":[162],"energy,":[166],"it":[167],"area":[171],"7.9%,":[173],"no":[175],"impact":[176],"or":[179],"compared":[181,210],"baseline":[184],"system.":[185],"Further,":[186],"show":[188],"partitioning-aware":[191],"pruning":[192],"strategy":[193,213],"EDP":[197],"49.7%":[204],"case":[207],"ResNeXt-50,":[209],"only":[215],"optimizes":[216],"individually.":[219],"Through":[220],"provided":[222],"information,":[223],"designers":[224],"receive":[225],"feedback":[226],"efficiency":[229],"full":[232],"at":[234],"early":[236],"development":[237],"stage.":[238],"Our":[239],"work":[240],"available":[242],"open":[243],"source1.1https://github.com/itiv-kit/cnn-parted":[244]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-12-09T00:00:00"}