{"id":"https://openalex.org/W4401608518","doi":"https://doi.org/10.1109/tpami.2024.3443916","title":"Safe Reinforcement Learning With Dual Robustness","display_name":"Safe Reinforcement Learning With Dual Robustness","publication_year":2024,"publication_date":"2024-08-15","ids":{"openalex":"https://openalex.org/W4401608518","doi":"https://doi.org/10.1109/tpami.2024.3443916","pmid":"https://pubmed.ncbi.nlm.nih.gov/39146157"},"language":"en","primary_location":{"id":"doi:10.1109/tpami.2024.3443916","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2024.3443916","pdf_url":null,"source":{"id":"https://openalex.org/S199944782","display_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","issn_l":"0162-8828","issn":["0162-8828","1939-3539","2160-9292"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320439","host_organization_name":"IEEE Computer Society","host_organization_lineage":["https://openalex.org/P4310320439","https://openalex.org/P4310319808"],"host_organization_lineage_names":["IEEE Computer Society","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 Pattern Analysis and Machine Intelligence","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/A5100779781","display_name":"Zeyang Li","orcid":"https://orcid.org/0009-0006-2868-1153"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Zeyang Li","raw_affiliation_strings":["Department of Mechanical Engineering, Tsinghua University, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0006-2868-1153","affiliations":[{"raw_affiliation_string":"Department of Mechanical Engineering, Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5079033650","display_name":"Chuxiong Hu","orcid":"https://orcid.org/0000-0002-3504-3065"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chuxiong Hu","raw_affiliation_strings":["Department of Mechanical Engineering, Tsinghua University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-3504-3065","affiliations":[{"raw_affiliation_string":"Department of Mechanical Engineering, Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5045375278","display_name":"Yunan Wang","orcid":"https://orcid.org/0009-0008-6065-7086"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yunan Wang","raw_affiliation_strings":["Department of Mechanical Engineering, Tsinghua University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-9812-7709","affiliations":[{"raw_affiliation_string":"Department of Mechanical Engineering, Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101620888","display_name":"Yujie Yang","orcid":"https://orcid.org/0000-0001-7222-0019"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yujie Yang","raw_affiliation_strings":["School of Vehicle and Mobility, Tsinghua University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0001-7222-0019","affiliations":[{"raw_affiliation_string":"School of Vehicle and Mobility, Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100747108","display_name":"Shengbo Eben Li","orcid":"https://orcid.org/0000-0003-4923-3633"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shengbo Eben Li","raw_affiliation_strings":["School of Vehicle and Mobility & College of AI, Tsinghua University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0003-4923-3633","affiliations":[{"raw_affiliation_string":"School of Vehicle and Mobility & College of AI, Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5100779781"],"corresponding_institution_ids":["https://openalex.org/I99065089"],"apc_list":null,"apc_paid":null,"fwci":3.9736,"has_fulltext":false,"cited_by_count":12,"citation_normalized_percentile":{"value":0.94386436,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":99},"biblio":{"volume":"46","issue":"12","first_page":"10876","last_page":"10890"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9961000084877014,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9961000084877014,"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":0.9814000129699707,"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/T13295","display_name":"Safety Systems Engineering in Autonomy","score":0.9726999998092651,"subfield":{"id":"https://openalex.org/subfields/2213","display_name":"Safety, Risk, Reliability and Quality"},"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/reinforcement-learning","display_name":"Reinforcement learning","score":0.8045989274978638},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6521174311637878},{"id":"https://openalex.org/keywords/markov-decision-process","display_name":"Markov decision process","score":0.6236710548400879},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.6070850491523743},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.603540301322937},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.4572136402130127},{"id":"https://openalex.org/keywords/q-learning","display_name":"Q-learning","score":0.435200035572052},{"id":"https://openalex.org/keywords/invariant","display_name":"Invariant (physics)","score":0.43418365716934204},{"id":"https://openalex.org/keywords/convergence","display_name":"Convergence (economics)","score":0.4171293377876282},{"id":"https://openalex.org/keywords/unification","display_name":"Unification","score":0.41653889417648315},{"id":"https://openalex.org/keywords/markov-process","display_name":"Markov process","score":0.372857928276062},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.36560577154159546},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.22817263007164001}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.8045989274978638},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6521174311637878},{"id":"https://openalex.org/C106189395","wikidata":"https://www.wikidata.org/wiki/Q176789","display_name":"Markov decision process","level":3,"score":0.6236710548400879},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.6070850491523743},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.603540301322937},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.4572136402130127},{"id":"https://openalex.org/C188116033","wikidata":"https://www.wikidata.org/wiki/Q2664563","display_name":"Q-learning","level":3,"score":0.435200035572052},{"id":"https://openalex.org/C190470478","wikidata":"https://www.wikidata.org/wiki/Q2370229","display_name":"Invariant (physics)","level":2,"score":0.43418365716934204},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.4171293377876282},{"id":"https://openalex.org/C96146094","wikidata":"https://www.wikidata.org/wiki/Q609057","display_name":"Unification","level":2,"score":0.41653889417648315},{"id":"https://openalex.org/C159886148","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov process","level":2,"score":0.372857928276062},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.36560577154159546},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.22817263007164001},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C50522688","wikidata":"https://www.wikidata.org/wiki/Q189833","display_name":"Economic growth","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0},{"id":"https://openalex.org/C37914503","wikidata":"https://www.wikidata.org/wiki/Q156495","display_name":"Mathematical physics","level":1,"score":0.0},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/tpami.2024.3443916","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2024.3443916","pdf_url":null,"source":{"id":"https://openalex.org/S199944782","display_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","issn_l":"0162-8828","issn":["0162-8828","1939-3539","2160-9292"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320439","host_organization_name":"IEEE Computer Society","host_organization_lineage":["https://openalex.org/P4310320439","https://openalex.org/P4310319808"],"host_organization_lineage_names":["IEEE Computer Society","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 Pattern Analysis and Machine Intelligence","raw_type":"journal-article"},{"id":"pmid:39146157","is_oa":false,"landing_page_url":"https://pubmed.ncbi.nlm.nih.gov/39146157","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 pattern analysis and machine intelligence","raw_type":null}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/10","score":0.4399999976158142,"display_name":"Reduced inequalities"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":47,"referenced_works":["https://openalex.org/W1845972764","https://openalex.org/W2106816694","https://openalex.org/W2145339207","https://openalex.org/W2158782408","https://openalex.org/W2160781031","https://openalex.org/W2492791808","https://openalex.org/W2560504659","https://openalex.org/W2736601468","https://openalex.org/W2766447205","https://openalex.org/W2911087563","https://openalex.org/W2968547875","https://openalex.org/W2998116579","https://openalex.org/W3011964880","https://openalex.org/W3013679533","https://openalex.org/W3110979110","https://openalex.org/W3162902207","https://openalex.org/W4206677550","https://openalex.org/W4221110788","https://openalex.org/W4224216677","https://openalex.org/W4226245271","https://openalex.org/W4309669389","https://openalex.org/W4311486598","https://openalex.org/W4317553716","https://openalex.org/W4353056919","https://openalex.org/W4362650413","https://openalex.org/W4364856217","https://openalex.org/W4367595613","https://openalex.org/W4379051467","https://openalex.org/W4388952994","https://openalex.org/W6638036514","https://openalex.org/W6684892534","https://openalex.org/W6687063787","https://openalex.org/W6735677848","https://openalex.org/W6737893269","https://openalex.org/W6741002519","https://openalex.org/W6747473740","https://openalex.org/W6751725685","https://openalex.org/W6759312646","https://openalex.org/W6771280675","https://openalex.org/W6774126978","https://openalex.org/W6774966973","https://openalex.org/W6785064911","https://openalex.org/W6804655846","https://openalex.org/W6810667139","https://openalex.org/W6838188116","https://openalex.org/W6838566501","https://openalex.org/W6904861994"],"related_works":["https://openalex.org/W1985560493","https://openalex.org/W2386410636","https://openalex.org/W2145363145","https://openalex.org/W1626977535","https://openalex.org/W4284974072","https://openalex.org/W2341346307","https://openalex.org/W4404955202","https://openalex.org/W3096874164","https://openalex.org/W2357975469","https://openalex.org/W4400868993"],"abstract_inverted_index":{"Reinforcement":[0],"learning":[1],"(RL)":[2],"agents":[3],"are":[4],"vulnerable":[5],"to":[6,66,102,116,164,184,215],"adversarial":[7,279],"disturbances,":[8],"which":[9,146,198,220,234],"can":[10],"deteriorate":[11],"task":[12,202,218,263,291],"performance":[13,40,320,331],"or":[14,34],"break":[15],"down":[16],"safety":[17,23,206,232,241,246,257,270,284,323,329],"specifications.":[18],"Existing":[19],"methods":[20],"either":[21],"address":[22,108],"requirements":[24],"under":[25,54,167,324],"the":[26,72,92,104,124,147,156,161,168,176,181,186,211,216,222,226,230,240,252,267,287],"assumption":[27],"of":[28,94,170,188,245,256,262,269],"no":[29],"adversary":[30],"(e.g.,":[31,42],"safe":[32,51,118],"RL)":[33],"only":[35,81],"focus":[36],"on":[37,100,266],"robustness":[38],"against":[39],"adversaries":[41,56],"robust":[43,53,88,121,158,178,307],"RL).":[44],"Learning":[45],"one":[46],"policy":[47,193,203,219,233,247,264],"that":[48,210,261,316],"is":[49,64,80,137,151,163,183,196,248,282,289],"both":[50],"and":[52,76,120,131,204,229,260,286,321],"any":[55],"remains":[57],"a":[58,84,113,201,205,295,337],"challenging":[59],"open":[60],"problem.":[61],"The":[62,78,135,243,310],"difficulty":[63],"how":[65,101],"tackle":[67],"two":[68,278],"intertwined":[69],"aspects":[70],"in":[71,145,225],"worst":[73,227],"cases:":[74],"feasibility":[75],"optimality.":[77],"optimality":[79],"valid":[82],"inside":[83,155],"feasible":[85,96],"region":[86,97],"(i.e.,":[87],"invariant":[89,159,179],"set),":[90],"while":[91],"identification":[93],"maximal":[95,157,177],"must":[98],"rely":[99],"learn":[103],"optimal":[105,217,231],"policy.":[106,207],"To":[107],"this":[109],"issue,":[110],"we":[111,293],"propose":[112,294],"systematic":[114],"framework":[115],"unify":[117],"RL":[119,298],"RL,":[122],"including":[123],"problem":[125],"formulation,":[126],"iteration":[127,194,212],"scheme,":[128],"convergence":[129,244],"analysis":[130],"practical":[132,296],"algorithm":[133,299],"design.":[134],"unification":[136],"built":[138],"upon":[139],"constrained":[140,301],"two-player":[141],"zero-sum":[142,302],"Markov":[143,303],"games,":[144,304],"objective":[148,224],"for":[149,173,283,290,300],"protagonist":[150],"twofold.":[152],"For":[153],"states":[154,174],"set,":[160,180],"goal":[162,182],"pursue":[165],"rewards":[166],"condition":[169],"guaranteed":[171],"safety;":[172],"outside":[175],"reduce":[185],"extent":[187],"constraint":[189],"violation.":[190],"A":[191],"dual":[192],"scheme":[195,213],"proposed,":[197],"simultaneously":[199],"optimizes":[200],"We":[208],"prove":[209],"converges":[214],"maximizes":[221],"twofold":[223],"cases,":[228],"stays":[235],"as":[236],"far":[237],"away":[238],"from":[239],"boundary.":[242],"established":[249],"by":[250,336],"exploiting":[251],"monotone":[253],"contraction":[254],"property":[255],"self-consistency":[258],"operators,":[259],"depends":[265],"transformation":[268],"constraints":[271],"into":[272],"state-dependent":[273],"action":[274],"spaces.":[275],"By":[276],"adding":[277],"networks":[280],"(one":[281],"guarantee":[285],"other":[288],"performance),":[292],"deep":[297],"called":[305],"dually":[306],"actor-critic":[308],"(DRAC).":[309],"evaluations":[311],"with":[312],"safety-critical":[313],"benchmarks":[314],"demonstrate":[315],"DRAC":[317],"achieves":[318],"high":[319],"persistent":[322],"all":[325,334],"scenarios":[326],"(no":[327],"adversary,":[328,330],"adversary),":[332],"outperforming":[333],"baselines":[335],"large":[338],"margin.":[339]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":8},{"year":2024,"cited_by_count":3}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}