{"id":"https://openalex.org/W7116918306","doi":"https://doi.org/10.1145/3772429.3772436","title":"SAC-B:Soft Actor-Critic with Bias for Suppressing Q-valueOverestimation in Off-Policy Reinforcement Learning","display_name":"SAC-B:Soft Actor-Critic with Bias for Suppressing Q-valueOverestimation in Off-Policy Reinforcement Learning","publication_year":2025,"publication_date":"2025-11-21","ids":{"openalex":"https://openalex.org/W7116918306","doi":"https://doi.org/10.1145/3772429.3772436"},"language":null,"primary_location":{"id":"doi:10.1145/3772429.3772436","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3772429.3772436","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2025 The Seventh International Conference on Distributed Artificial Intelligence","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3772429.3772436","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Han Wang","orcid":"https://orcid.org/0009-0002-3869-3941"},"institutions":[{"id":"https://openalex.org/I154099455","display_name":"Shandong University","ror":"https://ror.org/0207yh398","country_code":"CN","type":"education","lineage":["https://openalex.org/I154099455"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Han Wang","raw_affiliation_strings":["School of Software, Shandong University, Jinan, Shandong, China"],"raw_orcid":"https://orcid.org/0009-0002-3869-3941","affiliations":[{"raw_affiliation_string":"School of Software, Shandong University, Jinan, Shandong, China","institution_ids":["https://openalex.org/I154099455"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Wei Du","orcid":"https://orcid.org/0000-0001-5356-2427"},"institutions":[{"id":"https://openalex.org/I154099455","display_name":"Shandong University","ror":"https://ror.org/0207yh398","country_code":"CN","type":"education","lineage":["https://openalex.org/I154099455"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wei Du","raw_affiliation_strings":["School of Software, Shandong University, Jinan, Shandong, China"],"raw_orcid":"https://orcid.org/0000-0001-5356-2427","affiliations":[{"raw_affiliation_string":"School of Software, Shandong University, Jinan, Shandong, China","institution_ids":["https://openalex.org/I154099455"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5036903821","display_name":"Yanyu Xu","orcid":"https://orcid.org/0000-0001-8926-7833"},"institutions":[{"id":"https://openalex.org/I154099455","display_name":"Shandong University","ror":"https://ror.org/0207yh398","country_code":"CN","type":"education","lineage":["https://openalex.org/I154099455"]},{"id":"https://openalex.org/I4210097214","display_name":"Shandong Management University","ror":"https://ror.org/00vzprm14","country_code":"CN","type":"education","lineage":["https://openalex.org/I4210097214"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yanyu Xu","raw_affiliation_strings":["School of Software, Shandong University, Jinan,China, Shandong, China"],"raw_orcid":"https://orcid.org/0000-0001-8926-7833","affiliations":[{"raw_affiliation_string":"School of Software, Shandong University, Jinan,China, Shandong, China","institution_ids":["https://openalex.org/I154099455","https://openalex.org/I4210097214"]}]},{"author_position":"last","author":{"id":null,"display_name":"Lizhen Cui","orcid":"https://orcid.org/0000-0003-4977-7577"},"institutions":[{"id":"https://openalex.org/I154099455","display_name":"Shandong University","ror":"https://ror.org/0207yh398","country_code":"CN","type":"education","lineage":["https://openalex.org/I154099455"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lizhen Cui","raw_affiliation_strings":["School of Software, Shandong University, Jinan, Shandong, China"],"raw_orcid":"https://orcid.org/0000-0003-4977-7577","affiliations":[{"raw_affiliation_string":"School of Software, Shandong University, Jinan, Shandong, China","institution_ids":["https://openalex.org/I154099455"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"58","last_page":"66"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.902400016784668,"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":0.902400016784668,"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/T12794","display_name":"Adaptive Dynamic Programming Control","score":0.04190000146627426,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.010499999858438969,"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/reinforcement-learning","display_name":"Reinforcement learning","score":0.743399977684021},{"id":"https://openalex.org/keywords/entropy","display_name":"Entropy (arrow of time)","score":0.5455999970436096},{"id":"https://openalex.org/keywords/temporal-difference-learning","display_name":"Temporal difference learning","score":0.5232999920845032},{"id":"https://openalex.org/keywords/stability","display_name":"Stability (learning theory)","score":0.49129998683929443},{"id":"https://openalex.org/keywords/bootstrapping","display_name":"Bootstrapping (finance)","score":0.4187999963760376},{"id":"https://openalex.org/keywords/function","display_name":"Function (biology)","score":0.3806999921798706}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.743399977684021},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6705999970436096},{"id":"https://openalex.org/C106301342","wikidata":"https://www.wikidata.org/wiki/Q4117933","display_name":"Entropy (arrow of time)","level":2,"score":0.5455999970436096},{"id":"https://openalex.org/C196340769","wikidata":"https://www.wikidata.org/wiki/Q7698910","display_name":"Temporal difference learning","level":3,"score":0.5232999920845032},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.49129998683929443},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4634000062942505},{"id":"https://openalex.org/C207609745","wikidata":"https://www.wikidata.org/wiki/Q4944086","display_name":"Bootstrapping (finance)","level":2,"score":0.4187999963760376},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.39329999685287476},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.3806999921798706},{"id":"https://openalex.org/C9679016","wikidata":"https://www.wikidata.org/wiki/Q1417473","display_name":"Principle of maximum entropy","level":2,"score":0.3564999997615814},{"id":"https://openalex.org/C47446073","wikidata":"https://www.wikidata.org/wiki/Q5165890","display_name":"Control theory (sociology)","level":3,"score":0.31049999594688416},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.2994000017642975},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.26600000262260437},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.26460000872612},{"id":"https://openalex.org/C61797465","wikidata":"https://www.wikidata.org/wiki/Q1188986","display_name":"Term (time)","level":2,"score":0.2531999945640564}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3772429.3772436","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3772429.3772436","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2025 The Seventh International Conference on Distributed Artificial Intelligence","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3772429.3772436","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3772429.3772436","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2025 The Seventh International Conference on Distributed Artificial Intelligence","raw_type":"proceedings-article"},"sustainable_development_goals":[{"score":0.8007818460464478,"display_name":"Peace, Justice and strong institutions","id":"https://metadata.un.org/sdg/16"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":4,"referenced_works":["https://openalex.org/W2739431754","https://openalex.org/W2902907165","https://openalex.org/W4404037587","https://openalex.org/W4406983280"],"related_works":[],"abstract_inverted_index":{"Soft":[0,51],"Actor-Critic":[1,52],"(SAC)":[2],"and":[3,39,109,162,167],"other":[4],"off-policy":[5],"reinforcement":[6],"learning":[7,124],"methods":[8],"have":[9],"emerged":[10],"as":[11,79],"powerful":[12],"tools":[13],"for":[14],"complex":[15],"dynamic":[16,75],"decision-making.":[17],"However,":[18],"they":[19],"are":[20],"susceptible":[21],"to":[22,42,60,83,137,174],"detrimental":[23],"Q-value":[24,62,86],"overestimation":[25,63],"bias":[26,31,76,98],"in":[27],"stochastic":[28],"environments.":[29],"This":[30],"critically":[32],"impairs":[33],"policy":[34],"optimization,":[35],"propagates":[36],"estimation":[37],"errors,":[38],"can":[40],"lead":[41],"suboptimal":[43],"policies.":[44],"To":[45],"address":[46],"this":[47,97],"challenge,":[48],"we":[49,89],"introduce":[50],"with":[53,125],"Bias":[54],"(SAC-B),":[55],"a":[56,74,80,115],"novel":[57],"framework":[58],"designed":[59],"mitigate":[61],"while":[64,143],"preserving":[65],"the":[66,102,145,150,165],"strengths":[67],"of":[68,149,170],"maximum":[69],"entropy":[70,126],"RL.":[71],"SAC-B":[72,113,136],"incorporates":[73],"compensator,":[77],"implemented":[78],"learnable":[81],"network,":[82],"heuristically":[84],"counter":[85],"overestimation.":[87],"Crucially,":[88],"employ":[90],"an":[91],"error-decoupled":[92],"stabilized":[93],"architecture":[94],"that":[95,119],"isolates":[96],"prediction":[99],"module":[100],"from":[101],"critic":[103],"networks,":[104],"thereby":[105],"eliminating":[106],"error":[107],"propagation":[108],"enhancing":[110],"stability.":[111],"Furthermore,":[112],"utilizes":[114],"hybrid":[116],"objective":[117],"function":[118],"strategically":[120],"balances":[121],"temporal":[122,140],"difference":[123],"regularization,":[127],"enabling":[128],"rapid":[129],"adaptation.":[130],"These":[131],"synergistic":[132],"components":[133],"collectively":[134],"enable":[135],"effectively":[138],"reduce":[139],"bootstrapping":[141],"errors":[142],"maintaining":[144],"desirable":[146],"exploration":[147],"properties":[148],"original":[151],"SAC":[152],"algorithm.":[153],"Experiments":[154],"across":[155],"six":[156],"continuous":[157],"control":[158],"benchmarks,":[159],"including":[160],"BipedalWalkerHardcore-v3":[161],"Quadruped,":[163],"validate":[164],"superiority":[166],"enhanced":[168],"stability":[169],"our":[171],"approach":[172],"compared":[173],"existing":[175],"methods.":[176]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-12-23T00:00:00"}