{"id":"https://openalex.org/W4410294957","doi":"https://doi.org/10.1109/tpds.2025.3566766","title":"An Acceleration Framework for Deep Reinforcement Learning Using Heterogeneous Systems","display_name":"An Acceleration Framework for Deep Reinforcement Learning Using Heterogeneous Systems","publication_year":2025,"publication_date":"2025-05-12","ids":{"openalex":"https://openalex.org/W4410294957","doi":"https://doi.org/10.1109/tpds.2025.3566766"},"language":"en","primary_location":{"id":"doi:10.1109/tpds.2025.3566766","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpds.2025.3566766","pdf_url":null,"source":{"id":"https://openalex.org/S97130795","display_name":"IEEE Transactions on Parallel and Distributed Systems","issn_l":"1045-9219","issn":["1045-9219","1558-2183","2161-9883"],"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 Parallel and Distributed 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/A5050528589","display_name":"Yuan Meng","orcid":"https://orcid.org/0000-0001-6468-8623"},"institutions":[{"id":"https://openalex.org/I1174212","display_name":"University of Southern California","ror":"https://ror.org/03taz7m60","country_code":"US","type":"education","lineage":["https://openalex.org/I1174212"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Yuan Meng","raw_affiliation_strings":["Ming Hsieh Department of Electrical and Computer Engineering, University of Southern California, Los Angeles, CA, USA"],"affiliations":[{"raw_affiliation_string":"Ming Hsieh Department of Electrical and Computer Engineering, University of Southern California, Los Angeles, CA, USA","institution_ids":["https://openalex.org/I1174212"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5044099321","display_name":"Mahesh A. Iyer","orcid":"https://orcid.org/0000-0002-1045-0019"},"institutions":[{"id":"https://openalex.org/I22433950","display_name":"Altera (United States)","ror":"https://ror.org/017b7j426","country_code":"US","type":"company","lineage":["https://openalex.org/I22433950"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Mahesh A. Iyer","raw_affiliation_strings":["Altera Corporation, San Jose, CA, USA"],"affiliations":[{"raw_affiliation_string":"Altera Corporation, San Jose, CA, USA","institution_ids":["https://openalex.org/I22433950"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5033166029","display_name":"Viktor K. Prasanna","orcid":"https://orcid.org/0000-0002-1609-8589"},"institutions":[{"id":"https://openalex.org/I1174212","display_name":"University of Southern California","ror":"https://ror.org/03taz7m60","country_code":"US","type":"education","lineage":["https://openalex.org/I1174212"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Viktor K. Prasanna","raw_affiliation_strings":["Ming Hsieh Department of Electrical and Computer Engineering, University of Southern California, Los Angeles, CA, USA"],"affiliations":[{"raw_affiliation_string":"Ming Hsieh Department of Electrical and Computer Engineering, University of Southern California, Los Angeles, CA, USA","institution_ids":["https://openalex.org/I1174212"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5050528589"],"corresponding_institution_ids":["https://openalex.org/I1174212"],"apc_list":null,"apc_paid":null,"fwci":7.0705,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.96381013,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":95,"max":98},"biblio":{"volume":"36","issue":"7","first_page":"1401","last_page":"1415"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.6610000133514404,"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":0.6610000133514404,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7851976156234741},{"id":"https://openalex.org/keywords/acceleration","display_name":"Acceleration","score":0.7494755983352661},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.7373003959655762},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.40805351734161377},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.3444421887397766},{"id":"https://openalex.org/keywords/human\u2013computer-interaction","display_name":"Human\u2013computer interaction","score":0.32312726974487305}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7851976156234741},{"id":"https://openalex.org/C117896860","wikidata":"https://www.wikidata.org/wiki/Q11376","display_name":"Acceleration","level":2,"score":0.7494755983352661},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.7373003959655762},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.40805351734161377},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.3444421887397766},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.32312726974487305},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C74650414","wikidata":"https://www.wikidata.org/wiki/Q11397","display_name":"Classical mechanics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tpds.2025.3566766","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpds.2025.3566766","pdf_url":null,"source":{"id":"https://openalex.org/S97130795","display_name":"IEEE Transactions on Parallel and Distributed Systems","issn_l":"1045-9219","issn":["1045-9219","1558-2183","2161-9883"],"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 Parallel and Distributed Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G6237556090","display_name":null,"funder_award_id":"CNS-2009057","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G7267405974","display_name":null,"funder_award_id":"SPX-2333009","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320307102","display_name":"Intel Corporation","ror":"https://ror.org/01ek73717"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":35,"referenced_works":["https://openalex.org/W195033972","https://openalex.org/W1994616650","https://openalex.org/W2149294210","https://openalex.org/W2162741763","https://openalex.org/W2550668432","https://openalex.org/W2604883922","https://openalex.org/W2761873684","https://openalex.org/W2785452945","https://openalex.org/W2931767035","https://openalex.org/W2973229164","https://openalex.org/W3035681682","https://openalex.org/W3086105743","https://openalex.org/W3132871189","https://openalex.org/W3184015517","https://openalex.org/W3187908965","https://openalex.org/W3207730694","https://openalex.org/W4226369037","https://openalex.org/W4229017035","https://openalex.org/W4319870657","https://openalex.org/W4362653492","https://openalex.org/W6636881020","https://openalex.org/W6637967152","https://openalex.org/W6687681856","https://openalex.org/W6704571135","https://openalex.org/W6746866138","https://openalex.org/W6747473740","https://openalex.org/W6748554570","https://openalex.org/W6763300512","https://openalex.org/W6766978945","https://openalex.org/W6786314967","https://openalex.org/W6804601995","https://openalex.org/W6843081473","https://openalex.org/W6844748936","https://openalex.org/W6858422200","https://openalex.org/W6891880838"],"related_works":["https://openalex.org/W4306904969","https://openalex.org/W2138720691","https://openalex.org/W4362501864","https://openalex.org/W4380318855","https://openalex.org/W3084456289","https://openalex.org/W2024136090","https://openalex.org/W4391331176","https://openalex.org/W2031695474","https://openalex.org/W2586732548","https://openalex.org/W2964765435"],"abstract_inverted_index":{"Deep":[0],"Reinforcement":[1],"Learning":[2],"(DRL)":[3],"is":[4],"vital":[5],"in":[6,45,55],"various":[7,100],"AI":[8],"applications.":[9],"DRL":[10,33,47,69,118,165],"algorithms":[11],"comprise":[12],"diverse":[13,176],"compute":[14],"primitives,":[15],"which":[16],"may":[17],"not":[18],"be":[19],"simultaneously":[20],"optimized":[21],"using":[22,152],"a":[23,36,64],"homogeneous":[24],"architecture.":[25],"However,":[26],"even":[27],"with":[28,161,175],"available":[29],"heterogeneous":[30,72,124,173],"architectures,":[31],"optimizing":[32],"performance":[34,200],"remains":[35],"challenge":[37],"due":[38],"to":[39,195],"the":[40,50,93,142,153],"complexity":[41],"of":[42,52,75,92,114,128,193],"design":[43,106],"space":[44],"parallelizing":[46],"primitives":[48],"and":[49,79,102,108,134,140,169,179,197],"variety":[51],"hardware":[53,177],"employed":[54],"modern":[56],"data":[57],"centers.":[58],"To":[59],"address":[60],"this,":[61],"we":[62],"introduce":[63],"framework":[65,158],"for":[66,149,187],"composing":[67],"parallel":[68],"systems":[70],"on":[71,131,171],"platforms":[73,174,189],"consisting":[74],"general-purpose":[76],"processors":[77,101],"(CPUs)":[78],"accelerators":[80],"(GPUs,":[81],"FPGAs).":[82],"Our":[83],"innovations":[84],"include:":[85],"1.":[86],"A":[87],"general":[88],"training":[89,139],"protocol":[90],"agnostic":[91],"underlying":[94],"hardware,":[95],"enabling":[96,112],"portable":[97],"implementations":[98,183],"across":[99,202],"accelerators.":[103],"2.":[104],"Efficient":[105],"exploration":[107],"automatic":[109],"task":[110],"placement":[111],"parallelization":[113],"tasks":[115],"within":[116],"each":[117],"primitive":[119],"over":[120],"one":[121],"or":[122],"multiple":[123],"devices.":[125],"3.":[126],"Incorporation":[127],"DRL-specific":[129],"optimizations":[130],"runtime":[132],"scheduling":[133],"resource":[135],"allocation,":[136],"facilitating":[137],"parallelized":[138],"enhancing":[141],"overall":[143],"system":[144],"performance.":[145],"4.":[146],"High-level":[147],"API":[148],"productive":[150],"development":[151],"framework.":[154],"We":[155],"showcase":[156],"our":[157],"through":[159],"experimentation":[160],"three":[162,172],"widely":[163],"used":[164],"algorithms,":[166],"DQN,":[167],"DDPG,":[168],"SAC,":[170],"characteristics":[178],"interconnections.":[180],"The":[181],"generated":[182],"outperform":[184],"state-of-the-art":[185],"libraries":[186],"CPU-GPU":[188],"by":[190],"throughput":[191],"improvements":[192],"up":[194],"2\u00d7,":[196],"1.7\u00d7":[198],"higher":[199],"portability":[201],"platforms.":[203]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":2}],"updated_date":"2026-04-09T08:11:56.329763","created_date":"2025-10-10T00:00:00"}
