{"id":"https://openalex.org/W4405022829","doi":"https://doi.org/10.1109/tsmc.2024.3505613","title":"On the Effectiveness of Regularization Methods for Soft Actor-Critic in Discrete-Action Domains","display_name":"On the Effectiveness of Regularization Methods for Soft Actor-Critic in Discrete-Action Domains","publication_year":2024,"publication_date":"2024-12-04","ids":{"openalex":"https://openalex.org/W4405022829","doi":"https://doi.org/10.1109/tsmc.2024.3505613"},"language":"en","primary_location":{"id":"doi:10.1109/tsmc.2024.3505613","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tsmc.2024.3505613","pdf_url":null,"source":{"id":"https://openalex.org/S4210209078","display_name":"IEEE Transactions on Systems Man and Cybernetics Systems","issn_l":"2168-2216","issn":["2168-2216","2168-2232"],"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 Systems, Man, and Cybernetics: 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/A5077010288","display_name":"Bang Giang Le","orcid":null},"institutions":[{"id":"https://openalex.org/I67868205","display_name":"VNU University of Science","ror":"https://ror.org/05w54hk79","country_code":"VN","type":"education","lineage":["https://openalex.org/I177233841","https://openalex.org/I67868205"]}],"countries":["VN"],"is_corresponding":true,"raw_author_name":"Bang Giang Le","raw_affiliation_strings":["Human-Machine Interaction Laboratory, VNU University of Engineering and Technology, Hanoi, Vietnam"],"raw_orcid":"https://orcid.org/0009-0001-4948-6713","affiliations":[{"raw_affiliation_string":"Human-Machine Interaction Laboratory, VNU University of Engineering and Technology, Hanoi, Vietnam","institution_ids":["https://openalex.org/I67868205"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5053935756","display_name":"Viet Cuong Ta","orcid":"https://orcid.org/0000-0001-8058-5915"},"institutions":[{"id":"https://openalex.org/I67868205","display_name":"VNU University of Science","ror":"https://ror.org/05w54hk79","country_code":"VN","type":"education","lineage":["https://openalex.org/I177233841","https://openalex.org/I67868205"]}],"countries":["VN"],"is_corresponding":false,"raw_author_name":"Viet Cuong Ta","raw_affiliation_strings":["Human-Machine Interaction Laboratory, VNU University of Engineering and Technology, Hanoi, Vietnam"],"raw_orcid":"https://orcid.org/0000-0001-8058-5915","affiliations":[{"raw_affiliation_string":"Human-Machine Interaction Laboratory, VNU University of Engineering and Technology, Hanoi, Vietnam","institution_ids":["https://openalex.org/I67868205"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5077010288"],"corresponding_institution_ids":["https://openalex.org/I67868205"],"apc_list":null,"apc_paid":null,"fwci":0.9934,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.81397152,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":98},"biblio":{"volume":"55","issue":"2","first_page":"1425","last_page":"1438"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11574","display_name":"Artificial Intelligence in Games","score":0.8417999744415283,"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/T11574","display_name":"Artificial Intelligence in Games","score":0.8417999744415283,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.7907999753952026,"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/regularization","display_name":"Regularization (linguistics)","score":0.6377747058868408},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.5302291512489319},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.44474610686302185},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.40587297081947327},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.39170631766319275},{"id":"https://openalex.org/keywords/applied-mathematics","display_name":"Applied mathematics","score":0.3653470277786255},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.2798621356487274},{"id":"https://openalex.org/keywords/physics","display_name":"Physics","score":0.12170252203941345}],"concepts":[{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.6377747058868408},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.5302291512489319},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.44474610686302185},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.40587297081947327},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.39170631766319275},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.3653470277786255},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2798621356487274},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.12170252203941345},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tsmc.2024.3505613","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tsmc.2024.3505613","pdf_url":null,"source":{"id":"https://openalex.org/S4210209078","display_name":"IEEE Transactions on Systems Man and Cybernetics Systems","issn_l":"2168-2216","issn":["2168-2216","2168-2232"],"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 Systems, Man, and Cybernetics: Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":43,"referenced_works":["https://openalex.org/W2021725020","https://openalex.org/W2145339207","https://openalex.org/W2158782408","https://openalex.org/W2746553466","https://openalex.org/W2904246096","https://openalex.org/W2963403143","https://openalex.org/W2963576857","https://openalex.org/W3124201714","https://openalex.org/W3187550742","https://openalex.org/W4307230233","https://openalex.org/W4378805020","https://openalex.org/W6636774829","https://openalex.org/W6638018090","https://openalex.org/W6684205842","https://openalex.org/W6692846177","https://openalex.org/W6713603661","https://openalex.org/W6730111887","https://openalex.org/W6734206676","https://openalex.org/W6734517396","https://openalex.org/W6736495777","https://openalex.org/W6740471745","https://openalex.org/W6740879895","https://openalex.org/W6741002519","https://openalex.org/W6747473740","https://openalex.org/W6748839928","https://openalex.org/W6756287877","https://openalex.org/W6757021410","https://openalex.org/W6757058172","https://openalex.org/W6758608932","https://openalex.org/W6759637383","https://openalex.org/W6767334564","https://openalex.org/W6767569084","https://openalex.org/W6769218391","https://openalex.org/W6779129966","https://openalex.org/W6781166419","https://openalex.org/W6788860121","https://openalex.org/W6798240405","https://openalex.org/W6802202268","https://openalex.org/W6804601995","https://openalex.org/W6811106317","https://openalex.org/W6843222607","https://openalex.org/W6857206393","https://openalex.org/W7055423279"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W1979597421","https://openalex.org/W2007980826","https://openalex.org/W2061531152","https://openalex.org/W3002753104","https://openalex.org/W2077600819","https://openalex.org/W2142036596","https://openalex.org/W2072657027","https://openalex.org/W2962838298","https://openalex.org/W2600246793"],"abstract_inverted_index":{"Soft":[0],"actor-critic":[1],"(SAC)":[2],"is":[3,32,134],"a":[4,16,22,40,95],"reinforcement":[5],"learning":[6,87],"algorithm":[7],"that":[8,46,148],"employs":[9],"the":[10,29,36,47,53,62,78,82,86,90,100,106,110,124,141,153],"maximum":[11],"entropy":[12,38,54,83,101],"framework":[13],"to":[14,34,61,76,98,136],"train":[15],"stochastic":[17,30,91,154],"policy.":[18,92],"This":[19],"work":[20],"examines":[21],"specific":[23],"failure":[24],"case":[25],"of":[26,49,74,81,89,109,156],"SAC":[27,50,126,157],"where":[28],"policy":[31,111,155],"trained":[33],"maximize":[35],"expected":[37],"from":[39],"sparse":[41],"reward":[42],"environment.":[43],"We":[44],"demonstrate":[45],"over-exploration":[48],"can":[51],"make":[52],"temperature":[55,84,102],"collapse,":[56],"followed":[57],"by":[58],"unstable":[59],"updates":[60,108],"actor.":[63],"Based":[64],"on":[65,85,117],"our":[66,120,149],"analyses,":[67],"we":[68],"introduce":[69],"Reg-SAC,":[70],"an":[71],"improved":[72],"version":[73,151],"SAC,":[75],"mitigate":[77],"detrimental":[79],"effects":[80],"stability":[88],"Reg-SAC":[93,122],"incorporates":[94],"clipping":[96],"value":[97],"prevent":[99],"collapse":[103],"and":[104],"regularizes":[105],"gradient":[107],"via":[112],"Kullback-Leibler":[113],"divergence.":[114],"Through":[115],"experiments":[116],"discrete":[118],"benchmarks,":[119],"proposed":[121],"outperforms":[123],"standard":[125],"in":[127,140,160],"spare-reward":[128],"grid":[129],"world":[130],"environments":[131],"while":[132],"it":[133],"able":[135],"maintain":[137],"competitive":[138],"performance":[139],"dense-reward":[142],"Atari":[143],"benchmark.":[144],"The":[145],"results":[146],"highlight":[147],"regularized":[150],"makes":[152],"more":[158],"stable":[159],"discrete-action":[161],"domains.":[162]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":2}],"updated_date":"2025-12-22T23:10:17.713674","created_date":"2025-10-10T00:00:00"}