{"id":"https://openalex.org/W4416010704","doi":"https://doi.org/10.1109/jssc.2025.3626988","title":"A 40-nm 209-TOPS/W Reinforcement Learning Processor With Full Speculation Exploitation and Inference-Training Parallel Processing","display_name":"A 40-nm 209-TOPS/W Reinforcement Learning Processor With Full Speculation Exploitation and Inference-Training Parallel Processing","publication_year":2025,"publication_date":"2025-11-07","ids":{"openalex":"https://openalex.org/W4416010704","doi":"https://doi.org/10.1109/jssc.2025.3626988"},"language":null,"primary_location":{"id":"doi:10.1109/jssc.2025.3626988","is_oa":false,"landing_page_url":"https://doi.org/10.1109/jssc.2025.3626988","pdf_url":null,"source":{"id":"https://openalex.org/S83637746","display_name":"IEEE Journal of Solid-State Circuits","issn_l":"0018-9200","issn":["0018-9200","1558-173X"],"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 Journal of Solid-State Circuits","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/A5110237427","display_name":"Shih Hao Chen","orcid":null},"institutions":[{"id":"https://openalex.org/I16733864","display_name":"National Taiwan University","ror":"https://ror.org/05bqach95","country_code":"TW","type":"education","lineage":["https://openalex.org/I16733864"]}],"countries":["TW"],"is_corresponding":true,"raw_author_name":"Shih-Hao Chen","raw_affiliation_strings":["Graduate Institute of Electronics Engineering, National Taiwan University, Taipei, Taiwan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Graduate Institute of Electronics Engineering, National Taiwan University, Taipei, Taiwan","institution_ids":["https://openalex.org/I16733864"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5069622483","display_name":"Ping-Sheng Wu","orcid":"https://orcid.org/0009-0007-7825-9599"},"institutions":[{"id":"https://openalex.org/I16733864","display_name":"National Taiwan University","ror":"https://ror.org/05bqach95","country_code":"TW","type":"education","lineage":["https://openalex.org/I16733864"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Ping-Sheng Wu","raw_affiliation_strings":["Graduate Institute of Electronics Engineering, National Taiwan University, Taipei, Taiwan"],"raw_orcid":"https://orcid.org/0009-0007-7825-9599","affiliations":[{"raw_affiliation_string":"Graduate Institute of Electronics Engineering, National Taiwan University, Taipei, Taiwan","institution_ids":["https://openalex.org/I16733864"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5117596504","display_name":"Brian Dean Soon","orcid":"https://orcid.org/0009-0004-6032-0708"},"institutions":[{"id":"https://openalex.org/I16733864","display_name":"National Taiwan University","ror":"https://ror.org/05bqach95","country_code":"TW","type":"education","lineage":["https://openalex.org/I16733864"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Brian Dean Soon","raw_affiliation_strings":["Graduate School of Advanced Technology, National Taiwan University, Taipei, Taiwan"],"raw_orcid":"https://orcid.org/0009-0004-6032-0708","affiliations":[{"raw_affiliation_string":"Graduate School of Advanced Technology, National Taiwan University, Taipei, Taiwan","institution_ids":["https://openalex.org/I16733864"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102176313","display_name":"Chao-Hung Chen","orcid":null},"institutions":[{"id":"https://openalex.org/I4210148468","display_name":"Industrial Technology Research Institute","ror":"https://ror.org/05szzwt63","country_code":"TW","type":"nonprofit","lineage":["https://openalex.org/I4210148468"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Chao-Hung Chen","raw_affiliation_strings":["Industrial Technology Research Institute, Hsinchu, Taiwan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Industrial Technology Research Institute, Hsinchu, Taiwan","institution_ids":["https://openalex.org/I4210148468"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100642145","display_name":"Chih\u2010Wei Liu","orcid":"https://orcid.org/0000-0002-3006-9856"},"institutions":[{"id":"https://openalex.org/I4210148468","display_name":"Industrial Technology Research Institute","ror":"https://ror.org/05szzwt63","country_code":"TW","type":"nonprofit","lineage":["https://openalex.org/I4210148468"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Chih-Wei Liu","raw_affiliation_strings":["Industrial Technology Research Institute, Hsinchu, Taiwan"],"raw_orcid":"https://orcid.org/0000-0002-3006-9856","affiliations":[{"raw_affiliation_string":"Industrial Technology Research Institute, Hsinchu, Taiwan","institution_ids":["https://openalex.org/I4210148468"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103419215","display_name":"Chun\u2010Lung Hsu","orcid":null},"institutions":[{"id":"https://openalex.org/I4210148468","display_name":"Industrial Technology Research Institute","ror":"https://ror.org/05szzwt63","country_code":"TW","type":"nonprofit","lineage":["https://openalex.org/I4210148468"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Chun-Lung Hsu","raw_affiliation_strings":["Industrial Technology Research Institute, Hsinchu, Taiwan"],"raw_orcid":"https://orcid.org/0009-0003-3051-2269","affiliations":[{"raw_affiliation_string":"Industrial Technology Research Institute, Hsinchu, Taiwan","institution_ids":["https://openalex.org/I4210148468"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101750729","display_name":"Chia\u2010Hsiang Yang","orcid":"https://orcid.org/0000-0003-1163-321X"},"institutions":[{"id":"https://openalex.org/I16733864","display_name":"National Taiwan University","ror":"https://ror.org/05bqach95","country_code":"TW","type":"education","lineage":["https://openalex.org/I16733864"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Chia-Hsiang Yang","raw_affiliation_strings":["Department of Electrical Engineering and the Graduate Institute of Electronics Engineering, National Taiwan University, Taipei, Taiwan"],"raw_orcid":"https://orcid.org/0000-0003-1163-321X","affiliations":[{"raw_affiliation_string":"Department of Electrical Engineering and the Graduate Institute of Electronics Engineering, National Taiwan University, Taipei, Taiwan","institution_ids":["https://openalex.org/I16733864"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5110237427"],"corresponding_institution_ids":["https://openalex.org/I16733864"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.16506876,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"61","issue":"3","first_page":"1001","last_page":"1012"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.5974000096321106,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.5974000096321106,"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/T10653","display_name":"Robot Manipulation and Learning","score":0.022199999541044235,"subfield":{"id":"https://openalex.org/subfields/2207","display_name":"Control and Systems Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10784","display_name":"Muscle activation and electromyography studies","score":0.0203000009059906,"subfield":{"id":"https://openalex.org/subfields/2204","display_name":"Biomedical Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/speculation","display_name":"Speculation","score":0.625},{"id":"https://openalex.org/keywords/static-random-access-memory","display_name":"Static random-access memory","score":0.5059000253677368},{"id":"https://openalex.org/keywords/encoding","display_name":"Encoding (memory)","score":0.4952999949455261},{"id":"https://openalex.org/keywords/transpose","display_name":"Transpose","score":0.4918000102043152},{"id":"https://openalex.org/keywords/scheme","display_name":"Scheme (mathematics)","score":0.46970000863075256},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.46239998936653137},{"id":"https://openalex.org/keywords/parallel-processing","display_name":"Parallel processing","score":0.45100000500679016},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.4262000024318695},{"id":"https://openalex.org/keywords/efficient-energy-use","display_name":"Efficient energy use","score":0.41839998960494995}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8041999936103821},{"id":"https://openalex.org/C47941915","wikidata":"https://www.wikidata.org/wiki/Q107885","display_name":"Speculation","level":2,"score":0.625},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5404999852180481},{"id":"https://openalex.org/C68043766","wikidata":"https://www.wikidata.org/wiki/Q267416","display_name":"Static random-access memory","level":2,"score":0.5059000253677368},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.4952999949455261},{"id":"https://openalex.org/C200106649","wikidata":"https://www.wikidata.org/wiki/Q223683","display_name":"Transpose","level":3,"score":0.4918000102043152},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.46970000863075256},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.46239998936653137},{"id":"https://openalex.org/C106515295","wikidata":"https://www.wikidata.org/wiki/Q26806595","display_name":"Parallel processing","level":2,"score":0.45100000500679016},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4262000024318695},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.41839998960494995},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.41440001130104065},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.3747999966144562},{"id":"https://openalex.org/C103088060","wikidata":"https://www.wikidata.org/wiki/Q1062839","display_name":"Error detection and correction","level":2,"score":0.36570000648498535},{"id":"https://openalex.org/C12455157","wikidata":"https://www.wikidata.org/wiki/Q7835331","display_name":"Transposition (logic)","level":2,"score":0.3409999907016754},{"id":"https://openalex.org/C27163531","wikidata":"https://www.wikidata.org/wiki/Q7454694","display_name":"Serial memory processing","level":2,"score":0.3319000005722046},{"id":"https://openalex.org/C47487241","wikidata":"https://www.wikidata.org/wiki/Q5227230","display_name":"Data access","level":2,"score":0.3230000138282776},{"id":"https://openalex.org/C138827492","wikidata":"https://www.wikidata.org/wiki/Q6661985","display_name":"Data processing","level":2,"score":0.3215000033378601},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.31310001015663147},{"id":"https://openalex.org/C101722063","wikidata":"https://www.wikidata.org/wiki/Q218825","display_name":"Random access","level":2,"score":0.31060001254081726},{"id":"https://openalex.org/C107598950","wikidata":"https://www.wikidata.org/wiki/Q259864","display_name":"Microarchitecture","level":2,"score":0.305400013923645},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.30219998955726624},{"id":"https://openalex.org/C48372109","wikidata":"https://www.wikidata.org/wiki/Q3913","display_name":"Binary number","level":2,"score":0.299699991941452},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.29840001463890076},{"id":"https://openalex.org/C17349429","wikidata":"https://www.wikidata.org/wiki/Q1049914","display_name":"Matrix multiplication","level":3,"score":0.2946000099182129},{"id":"https://openalex.org/C176649486","wikidata":"https://www.wikidata.org/wiki/Q2308807","display_name":"Memory management","level":3,"score":0.28949999809265137},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.2750999927520752},{"id":"https://openalex.org/C106487976","wikidata":"https://www.wikidata.org/wiki/Q685816","display_name":"Matrix (chemical analysis)","level":2,"score":0.2587999999523163},{"id":"https://openalex.org/C126042441","wikidata":"https://www.wikidata.org/wiki/Q1324888","display_name":"Frame (networking)","level":2,"score":0.2567000091075897},{"id":"https://openalex.org/C2780490138","wikidata":"https://www.wikidata.org/wiki/Q7079636","display_name":"Offline learning","level":3,"score":0.2554999887943268},{"id":"https://openalex.org/C117280010","wikidata":"https://www.wikidata.org/wiki/Q180944","display_name":"Register file","level":3,"score":0.25529998540878296},{"id":"https://openalex.org/C75684735","wikidata":"https://www.wikidata.org/wiki/Q858810","display_name":"Big data","level":2,"score":0.251800000667572}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/jssc.2025.3626988","is_oa":false,"landing_page_url":"https://doi.org/10.1109/jssc.2025.3626988","pdf_url":null,"source":{"id":"https://openalex.org/S83637746","display_name":"IEEE Journal of Solid-State Circuits","issn_l":"0018-9200","issn":["0018-9200","1558-173X"],"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 Journal of Solid-State Circuits","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320322214","display_name":"Industrial Technology Research Institute","ror":"https://ror.org/05szzwt63"},{"id":"https://openalex.org/F4320331164","display_name":"National Science and Technology Council","ror":"https://ror.org/00wnb9798"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":14,"referenced_works":["https://openalex.org/W2107726111","https://openalex.org/W2766447205","https://openalex.org/W2904246096","https://openalex.org/W2920954974","https://openalex.org/W2965347706","https://openalex.org/W2968983352","https://openalex.org/W2982316857","https://openalex.org/W3015729306","https://openalex.org/W3126321819","https://openalex.org/W3183385172","https://openalex.org/W3206827162","https://openalex.org/W4221141531","https://openalex.org/W4286571873","https://openalex.org/W4410492703"],"related_works":[],"abstract_inverted_index":{"Reinforcement":[0],"learning":[1],"(RL)":[2],"has":[3],"found":[4],"widespread":[5],"applications":[6],"across":[7],"diverse":[8],"domains,":[9],"making":[10],"energy-efficient":[11,18],"implementations":[12],"imperative.":[13],"This":[14,193],"article":[15],"presents":[16],"an":[17,178,187],"RL":[19,175],"processor":[20,118,176],"featuring":[21],"full":[22,52,106],"speculation":[23,53,111],"exploitation":[24],"and":[25,30,66,85,110,125,163,186,203,215],"parallel":[26],"processing":[27],"for":[28,61,108,143,154,161],"inference":[29,124],"training.":[31],"Binary":[32],"direct":[33],"feedback":[34],"alignment":[35],"(DFA)":[36],"is":[37,55,72,89,100,119,133],"applied":[38],"to":[39,57,91,102,122,135],"perform":[40,103,123],"error":[41,64],"propagation":[42],"in":[43,127,170,213],"parallel,":[44],"reducing":[45],"the":[46,59,93,149,173,219],"computational":[47,70],"complexity":[48,71],"by":[49,75,96,146,167],"23%.":[50],"A":[51,129],"scheme":[54,81,132],"proposed":[56,90,174],"estimate":[58],"sparsity":[60,109],"forward":[62],"propagation,":[63,65],"gradient":[67],"calculation.":[68],"The":[69,98,117],"further":[73],"reduced":[74],"87%.":[76],"An":[77],"efficient":[78],"data":[79,87,156],"encoding":[80],"leveraging":[82],"both":[83],"spatial":[84],"temporal":[86],"correlations":[88],"reduce":[92,136],"bitmask":[94],"size":[95],"65%.":[97],"architecture":[99],"designed":[101,121],"operations":[104],"with":[105],"support":[107],"while":[112,148],"maintaining":[113],"high":[114],"hardware":[115],"utilization.":[116],"also":[120],"training":[126],"parallel.":[128],"block-based":[130],"transposition":[131],"used":[134],"static":[137],"random":[138],"access":[139,142,160],"memory":[140],"(SRAM)":[141],"matrix":[144],"transpose":[145],"39%,":[147],"shared":[150],"input/output":[151],"(I/O)":[152],"buffer":[153],"internal":[155],"casting":[157],"reduces":[158],"SRAM":[159],"input":[162],"output":[164],"features":[165],"movement":[166],"35%.":[168],"Fabricated":[169],"40-nm":[171],"CMOS,":[172],"delivers":[177],"area":[179,214],"efficiency":[180,189,211],"of":[181,190],"2341":[182],"GOPS/mm<sup":[183],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[184,197,205],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">2</sup>":[185],"energy":[188],"209":[191],"TOPS/W.":[192],"work":[194],"achieves":[195],"<inline-formula":[196,204],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">":[198,206],"<tex-math":[199,207],"notation=\"LaTeX\">$7.3{\\times":[200],"}$</tex-math>":[201,209],"</inline-formula>":[202,210],"notation=\"LaTeX\">$7.1{\\times":[208],"improvements":[212],"energy,":[216],"respectively,":[217],"over":[218],"state-of-the-art":[220],"designs.":[221]},"counts_by_year":[],"updated_date":"2026-05-05T08:41:31.759640","created_date":"2025-11-07T00:00:00"}