{"id":"https://openalex.org/W4404133862","doi":"https://doi.org/10.1145/3649329.3657369","title":"SPARK: An Efficient Hybrid Acceleration Architecture with Run-Time Sparsity-Aware Scheduling for TinyML Learning","display_name":"SPARK: An Efficient Hybrid Acceleration Architecture with Run-Time Sparsity-Aware Scheduling for TinyML Learning","publication_year":2024,"publication_date":"2024-06-23","ids":{"openalex":"https://openalex.org/W4404133862","doi":"https://doi.org/10.1145/3649329.3657369"},"language":"en","primary_location":{"id":"doi:10.1145/3649329.3657369","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3649329.3657369","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 61st ACM/IEEE Design Automation Conference","raw_type":"proceedings-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/A5100365307","display_name":"Mingxuan Li","orcid":"https://orcid.org/0009-0008-1235-147X"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Mingxuan Li","raw_affiliation_strings":["School of Integrated Circuits, Peking University, Beijing, Beijing, China"],"affiliations":[{"raw_affiliation_string":"School of Integrated Circuits, Peking University, Beijing, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114554075","display_name":"Qinzhe Zhi","orcid":"https://orcid.org/0009-0001-4866-7213"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qinzhe Zhi","raw_affiliation_strings":["School of Integrated Circuits, Peking University, Beijing, Beijing, China"],"affiliations":[{"raw_affiliation_string":"School of Integrated Circuits, Peking University, Beijing, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5085097472","display_name":"Yanchi Dong","orcid":"https://orcid.org/0009-0004-5321-8297"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yanchi Dong","raw_affiliation_strings":["School of Integrated Circuits, Peking University, Beijing, Beijing, China"],"affiliations":[{"raw_affiliation_string":"School of Integrated Circuits, Peking University, Beijing, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003039083","display_name":"Le Ye","orcid":"https://orcid.org/0000-0003-0599-7762"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Le Ye","raw_affiliation_strings":["Advanced Institute of Information Technology of Peking University, Hangzhou, Zhejiang, China","School of Integrated Circuits, Peking University, Beijing, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Advanced Institute of Information Technology of Peking University, Hangzhou, Zhejiang, China","institution_ids":["https://openalex.org/I20231570"]},{"raw_affiliation_string":"School of Integrated Circuits, Peking University, Beijing, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5088551028","display_name":"Tianyu Jia","orcid":"https://orcid.org/0000-0002-4570-4613"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tianyu Jia","raw_affiliation_strings":["School of Integrated Circuits, Peking University, Beijing, Beijing, China"],"affiliations":[{"raw_affiliation_string":"School of Integrated Circuits, Peking University, Beijing, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5100365307"],"corresponding_institution_ids":["https://openalex.org/I20231570"],"apc_list":null,"apc_paid":null,"fwci":0.5244,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.67465804,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"6"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9991999864578247,"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":0.9991999864578247,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9991000294685364,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9977999925613403,"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/computer-science","display_name":"Computer science","score":0.7676078677177429},{"id":"https://openalex.org/keywords/spark","display_name":"SPARK (programming language)","score":0.6973251700401306},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.6113414168357849},{"id":"https://openalex.org/keywords/architecture","display_name":"Architecture","score":0.5982778072357178},{"id":"https://openalex.org/keywords/acceleration","display_name":"Acceleration","score":0.5788536667823792},{"id":"https://openalex.org/keywords/parallel-architecture","display_name":"Parallel architecture","score":0.46455925703048706},{"id":"https://openalex.org/keywords/job-shop-scheduling","display_name":"Job shop scheduling","score":0.45025694370269775},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.4093695282936096},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.3886246085166931},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.38425374031066895},{"id":"https://openalex.org/keywords/real-time-computing","display_name":"Real-time computing","score":0.35534989833831787},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.2613401412963867},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.10193419456481934}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7676078677177429},{"id":"https://openalex.org/C2781215313","wikidata":"https://www.wikidata.org/wiki/Q3493345","display_name":"SPARK (programming language)","level":2,"score":0.6973251700401306},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.6113414168357849},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.5982778072357178},{"id":"https://openalex.org/C117896860","wikidata":"https://www.wikidata.org/wiki/Q11376","display_name":"Acceleration","level":2,"score":0.5788536667823792},{"id":"https://openalex.org/C2985918086","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel architecture","level":3,"score":0.46455925703048706},{"id":"https://openalex.org/C55416958","wikidata":"https://www.wikidata.org/wiki/Q6206757","display_name":"Job shop scheduling","level":3,"score":0.45025694370269775},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4093695282936096},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.3886246085166931},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.38425374031066895},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.35534989833831787},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.2613401412963867},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.10193419456481934},{"id":"https://openalex.org/C74172769","wikidata":"https://www.wikidata.org/wiki/Q1446839","display_name":"Routing (electronic design automation)","level":2,"score":0.0},{"id":"https://openalex.org/C74650414","wikidata":"https://www.wikidata.org/wiki/Q11397","display_name":"Classical mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C21547014","wikidata":"https://www.wikidata.org/wiki/Q1423657","display_name":"Operations management","level":1,"score":0.0},{"id":"https://openalex.org/C153349607","wikidata":"https://www.wikidata.org/wiki/Q36649","display_name":"Visual arts","level":1,"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/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0},{"id":"https://openalex.org/C142362112","wikidata":"https://www.wikidata.org/wiki/Q735","display_name":"Art","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3649329.3657369","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3649329.3657369","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 61st ACM/IEEE Design Automation Conference","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.7900000214576721,"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":15,"referenced_works":["https://openalex.org/W2165698076","https://openalex.org/W2289252105","https://openalex.org/W2764043458","https://openalex.org/W2796438033","https://openalex.org/W2902251695","https://openalex.org/W2912213068","https://openalex.org/W2921918777","https://openalex.org/W3005710104","https://openalex.org/W3015729306","https://openalex.org/W3135461891","https://openalex.org/W3207686785","https://openalex.org/W4221143038","https://openalex.org/W4386764085","https://openalex.org/W6770699648","https://openalex.org/W6838539104"],"related_works":["https://openalex.org/W1975949872","https://openalex.org/W3159871278","https://openalex.org/W2230552005","https://openalex.org/W2905242764","https://openalex.org/W3109411864","https://openalex.org/W3017846737","https://openalex.org/W3172150420","https://openalex.org/W2158228144","https://openalex.org/W206792560","https://openalex.org/W1965117583"],"abstract_inverted_index":{"Currently":[0],"most":[1],"TinyML":[2,31,101],"devices":[3,125],"only":[4,116],"focus":[5],"on":[6],"inference,":[7],"as":[8],"training":[9],"requires":[10],"much":[11],"more":[12],"hardware":[13,60],"resources.":[14],"In":[15],"this":[16],"paper,":[17],"we":[18],"introduce":[19],"SPARK,":[20],"an":[21,37],"efficient":[22],"hybrid":[23],"acceleration":[24,39,64,74],"architecture":[25],"with":[26,104,115,130],"run-time":[27],"sparsity-aware":[28,63],"scheduling":[29],"for":[30],"learning.":[32],"Besides":[33],"a":[34,62],"standalone":[35],"accelerator,":[36,107],"in-pipeline":[38],"unit":[40],"is":[41,66,80],"integrated":[42],"within":[43],"the":[44,70,105],"CPU":[45],"pipeline":[46],"to":[47,68,83],"support":[48,84],"simultaneous":[49],"forward":[50],"and":[51,58,98],"backward":[52],"propagation.":[53],"To":[54],"better":[55],"utilize":[56],"sparsity":[57],"improve":[59],"utilization,":[61],"scheduler":[65],"implemented":[67],"schedule":[69],"workload":[71],"between":[72],"two":[73],"units.":[75],"A":[76],"unified":[77],"memory":[78,89],"system":[79],"also":[81,121],"constructed":[82],"transposable":[85],"data":[86],"fetch,":[87],"reducing":[88],"access.":[90],"We":[91],"implement":[92],"SPARK":[93,108,120],"using":[94],"TSMC":[95],"22nm":[96],"technology":[97],"evaluate":[99],"different":[100],"tasks.":[102],"Compared":[103],"baseline":[106],"achieves":[109],"4.1\u00d7":[110],"performance":[111,127],"improvement":[112],"in":[113,126],"average":[114],"2.27%":[117],"area":[118],"overhead.":[119],"outperforms":[122],"off-shelf":[123],"edge":[124],"by":[128],"9.4\u00d7":[129],"446.0\u00d7":[131],"higher":[132],"efficiency.":[133]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1}],"updated_date":"2026-03-06T13:50:29.536080","created_date":"2025-10-10T00:00:00"}