{"id":"https://openalex.org/W4417508524","doi":"https://doi.org/10.1109/tnnls.2025.3640101","title":"The Most Overestimated <i>Q</i> Value Regularization in High-Dimensional Discrete Action Spaces for Offline Reinforcement Learning","display_name":"The Most Overestimated <i>Q</i> Value Regularization in High-Dimensional Discrete Action Spaces for Offline Reinforcement Learning","publication_year":2025,"publication_date":"2025-12-19","ids":{"openalex":"https://openalex.org/W4417508524","doi":"https://doi.org/10.1109/tnnls.2025.3640101","pmid":"https://pubmed.ncbi.nlm.nih.gov/41418013"},"language":"en","primary_location":{"id":"doi:10.1109/tnnls.2025.3640101","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tnnls.2025.3640101","pdf_url":null,"source":{"id":"https://openalex.org/S4210175523","display_name":"IEEE Transactions on Neural Networks and Learning Systems","issn_l":"2162-237X","issn":["2162-237X","2162-2388"],"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 Neural Networks and Learning Systems","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","pubmed"],"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/A5113745386","display_name":"S. Yu","orcid":"https://orcid.org/0000-0002-0851-7392"},"institutions":[{"id":"https://openalex.org/I4575257","display_name":"Hanyang University","ror":"https://ror.org/046865y68","country_code":"KR","type":"education","lineage":["https://openalex.org/I4575257"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Seunghwan Yu","raw_affiliation_strings":["Department of Applied Artificial Intelligence Engineering, Hanyang University, Seoul, South Korea"],"raw_orcid":"https://orcid.org/0000-0002-0851-7392","affiliations":[{"raw_affiliation_string":"Department of Applied Artificial Intelligence Engineering, Hanyang University, Seoul, South Korea","institution_ids":["https://openalex.org/I4575257"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5081670766","display_name":"Ho-min Park","orcid":"https://orcid.org/0000-0001-9937-8617"},"institutions":[{"id":"https://openalex.org/I8696757","display_name":"Singapore University of Social Sciences","ror":"https://ror.org/01s57k749","country_code":"SG","type":"education","lineage":["https://openalex.org/I8696757"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Homin Park","raw_affiliation_strings":["School of Science and Technology, Singapore University of Social Sciences, Singapore, Singapore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Science and Technology, Singapore University of Social Sciences, Singapore, Singapore","institution_ids":["https://openalex.org/I8696757"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5050742077","display_name":"Byungjin Ko","orcid":"https://orcid.org/0000-0003-3786-9896"},"institutions":[{"id":"https://openalex.org/I4575257","display_name":"Hanyang University","ror":"https://ror.org/046865y68","country_code":"KR","type":"education","lineage":["https://openalex.org/I4575257"]},{"id":"https://openalex.org/I93906172","display_name":"Anyang University","ror":"https://ror.org/018pdh902","country_code":"KR","type":"education","lineage":["https://openalex.org/I93906172"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Byungjin Ko","raw_affiliation_strings":["Division of Smart Convergence Engineering, Hanyang University ERICA Campus, Ansan, Gyeonggi-do, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Division of Smart Convergence Engineering, Hanyang University ERICA Campus, Ansan, Gyeonggi-do, South Korea","institution_ids":["https://openalex.org/I4575257","https://openalex.org/I93906172"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Jisub Shin","orcid":null},"institutions":[{"id":"https://openalex.org/I4575257","display_name":"Hanyang University","ror":"https://ror.org/046865y68","country_code":"KR","type":"education","lineage":["https://openalex.org/I4575257"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jisub Shin","raw_affiliation_strings":["Department of Applied Artificial Intelligence Engineering, Hanyang University, Seoul, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Applied Artificial Intelligence Engineering, Hanyang University, Seoul, South Korea","institution_ids":["https://openalex.org/I4575257"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5108226543","display_name":"Y. S. Hong","orcid":"https://orcid.org/0009-0003-4704-3913"},"institutions":[{"id":"https://openalex.org/I4575257","display_name":"Hanyang University","ror":"https://ror.org/046865y68","country_code":"KR","type":"education","lineage":["https://openalex.org/I4575257"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Yoonki Hong","raw_affiliation_strings":["Department of Applied Artificial Intelligence Engineering, Hanyang University, Seoul, South Korea"],"raw_orcid":"https://orcid.org/0009-0003-4704-3913","affiliations":[{"raw_affiliation_string":"Department of Applied Artificial Intelligence Engineering, Hanyang University, Seoul, South Korea","institution_ids":["https://openalex.org/I4575257"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101586319","display_name":"Taejoon Park","orcid":"https://orcid.org/0000-0002-7924-1776"},"institutions":[{"id":"https://openalex.org/I4575257","display_name":"Hanyang University","ror":"https://ror.org/046865y68","country_code":"KR","type":"education","lineage":["https://openalex.org/I4575257"]},{"id":"https://openalex.org/I93906172","display_name":"Anyang University","ror":"https://ror.org/018pdh902","country_code":"KR","type":"education","lineage":["https://openalex.org/I93906172"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Taejoon Park","raw_affiliation_strings":["Department of Robotics Engineering, Hanyang University ERICA Campus, Ansan, Gyeonggi-do, South Korea"],"raw_orcid":"https://orcid.org/0000-0002-7924-1776","affiliations":[{"raw_affiliation_string":"Department of Robotics Engineering, Hanyang University ERICA Campus, Ansan, Gyeonggi-do, South Korea","institution_ids":["https://openalex.org/I4575257","https://openalex.org/I93906172"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5069766217","display_name":"Jong\u2010Won Yoon","orcid":"https://orcid.org/0000-0001-8978-9113"},"institutions":[{"id":"https://openalex.org/I4575257","display_name":"Hanyang University","ror":"https://ror.org/046865y68","country_code":"KR","type":"education","lineage":["https://openalex.org/I4575257"]},{"id":"https://openalex.org/I93906172","display_name":"Anyang University","ror":"https://ror.org/018pdh902","country_code":"KR","type":"education","lineage":["https://openalex.org/I93906172"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jong-Wan Yoon","raw_affiliation_strings":["Department of Intelligent Robotics, Hanyang University ERICA Campus, Ansan, Gyeonggi-do, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Intelligent Robotics, Hanyang University ERICA Campus, Ansan, Gyeonggi-do, South Korea","institution_ids":["https://openalex.org/I4575257","https://openalex.org/I93906172"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":7,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.7092,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.89733967,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":98},"biblio":{"volume":"37","issue":"6","first_page":"2794","last_page":"2808"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.6949999928474426,"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":0.6949999928474426,"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":0.23960000276565552,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.010599999688565731,"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":0.8269000053405762},{"id":"https://openalex.org/keywords/regularization","display_name":"Regularization (linguistics)","score":0.7339000105857849},{"id":"https://openalex.org/keywords/adaptability","display_name":"Adaptability","score":0.570900022983551},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.5077000260353088},{"id":"https://openalex.org/keywords/robotics","display_name":"Robotics","score":0.41530001163482666},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.4090999960899353},{"id":"https://openalex.org/keywords/value","display_name":"Value (mathematics)","score":0.3564000129699707},{"id":"https://openalex.org/keywords/flexibility","display_name":"Flexibility (engineering)","score":0.3409999907016754}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.8269000053405762},{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.7339000105857849},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6251999735832214},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6157000064849854},{"id":"https://openalex.org/C177606310","wikidata":"https://www.wikidata.org/wiki/Q5674297","display_name":"Adaptability","level":2,"score":0.570900022983551},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.5077000260353088},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4440000057220459},{"id":"https://openalex.org/C34413123","wikidata":"https://www.wikidata.org/wiki/Q170978","display_name":"Robotics","level":3,"score":0.41530001163482666},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.4090999960899353},{"id":"https://openalex.org/C2776291640","wikidata":"https://www.wikidata.org/wiki/Q2912517","display_name":"Value (mathematics)","level":2,"score":0.3564000129699707},{"id":"https://openalex.org/C2780598303","wikidata":"https://www.wikidata.org/wiki/Q65921492","display_name":"Flexibility (engineering)","level":2,"score":0.3409999907016754},{"id":"https://openalex.org/C188116033","wikidata":"https://www.wikidata.org/wiki/Q2664563","display_name":"Q-learning","level":3,"score":0.33660000562667847},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.32580000162124634},{"id":"https://openalex.org/C2780490138","wikidata":"https://www.wikidata.org/wiki/Q7079636","display_name":"Offline learning","level":3,"score":0.3125999867916107},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.3068999946117401},{"id":"https://openalex.org/C150415221","wikidata":"https://www.wikidata.org/wiki/Q40687","display_name":"Robotic arm","level":2,"score":0.2946000099182129},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.28949999809265137},{"id":"https://openalex.org/C2986087404","wikidata":"https://www.wikidata.org/wiki/Q15946010","display_name":"Online learning","level":2,"score":0.25769999623298645},{"id":"https://openalex.org/C2987834672","wikidata":"https://www.wikidata.org/wiki/Q4677630","display_name":"Action recognition","level":3,"score":0.25519999861717224},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.2508000135421753}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/tnnls.2025.3640101","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tnnls.2025.3640101","pdf_url":null,"source":{"id":"https://openalex.org/S4210175523","display_name":"IEEE Transactions on Neural Networks and Learning Systems","issn_l":"2162-237X","issn":["2162-237X","2162-2388"],"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 Neural Networks and Learning Systems","raw_type":"journal-article"},{"id":"pmid:41418013","is_oa":false,"landing_page_url":"https://pubmed.ncbi.nlm.nih.gov/41418013","pdf_url":null,"source":{"id":"https://openalex.org/S4306525036","display_name":"PubMed","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1299303238","host_organization_name":"National Institutes of Health","host_organization_lineage":["https://openalex.org/I1299303238"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE transactions on neural networks and learning systems","raw_type":null}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Deep":[0],"reinforcement":[1,32],"learning":[2,6,33],"(DRL)":[3],"excels":[4],"at":[5,224],"control":[7],"policies":[8],"in":[9,50,98,124,152,180,184,212,218],"high-dimensional":[10,51,99,147],"action":[11,53,88,101,106,121,149,191],"spaces,":[12,54],"making":[13],"it":[14],"crucial":[15],"for":[16,203,209],"robotic":[17,140,204],"manipulation.":[18],"However,":[19],"its":[20,207],"real-world":[21,156,185,210],"application":[22],"is":[23,221],"limited":[24],"by":[25,39,109],"costly":[26],"and":[27,142,155,161,182,193,201],"risky":[28],"data":[29],"collection.":[30],"Offline":[31],"(offline":[34],"RL)":[35],"addresses":[36],"this":[37,69],"issue":[38],"training":[40,66],"on":[41,138],"precollected":[42],"datasets":[43],"but":[44],"struggles":[45],"with":[46,89,158],"$Q$":[47,76,93,110],"value":[48,77,111],"overestimation":[49,97],"discrete":[52,100,148],"where":[55],"the":[56,73,87,90,105,119,130],"number":[57],"of":[58,178],"out-of-distribution":[59],"(OOD)":[60],"actions":[61],"rapidly":[62],"increases,":[63],"negatively":[64],"impacting":[65],"stability.":[67,194],"In":[68],"work,":[70],"we":[71],"propose":[72],"most":[74,91,107],"overestimated":[75,92],"regularization":[78],"(MQR),":[79],"a":[80,139,145,175],"novel":[81],"offline":[82],"RL":[83],"algorithm":[84],"that":[85,168],"penalizes":[86],"value,":[94],"effectively":[95],"mitigating":[96],"spaces.":[102],"By":[103],"regulating":[104],"affected":[108],"overestimation,":[112],"rather":[113],"than":[114],"applying":[115],"uniform":[116],"penalties":[117],"across":[118],"entire":[120],"space":[122,150],"as":[123],"existing":[125],"methods,":[126],"MQR":[127,137,169],"further":[128],"prevents":[129],"policy":[131],"from":[132],"converging":[133],"incorrectly.":[134],"We":[135],"evaluate":[136],"pushing":[141],"grasping":[143],"task,":[144],"challenging":[146],"problem,":[151],"both":[153],"simulated":[154],"environments":[157],"random,":[159],"dense,":[160],"unknown":[162],"object":[163],"arrangements.":[164],"The":[165,215],"results":[166],"demonstrate":[167],"significantly":[170],"outperforms":[171],"baseline":[172],"algorithms,":[173],"achieving":[174],"clearance":[176],"rate":[177],"96.94%":[179],"simulations":[181],"99.04%":[183],"dense":[186],"configurations,":[187],"while":[188],"maintaining":[189],"high":[190],"efficiency":[192],"These":[195],"findings":[196],"highlight":[197],"MQR's":[198],"robustness,":[199],"scalability,":[200],"adaptability":[202],"manipulation,":[205],"showcasing":[206],"potential":[208],"deployment":[211],"industrial":[213],"robotics.":[214],"code":[216],"used":[217],"our":[219],"research":[220],"publicly":[222],"available":[223],"https://github.com/Hanyang-Robot/MQR.":[225]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-06-17T08:01:34.144755","created_date":"2025-12-19T00:00:00"}