{"id":"https://openalex.org/W4391021813","doi":"https://doi.org/10.1109/cdc49753.2023.10383375","title":"Robust Safe Reinforcement Learning under Adversarial Disturbances","display_name":"Robust Safe Reinforcement Learning under Adversarial Disturbances","publication_year":2023,"publication_date":"2023-12-13","ids":{"openalex":"https://openalex.org/W4391021813","doi":"https://doi.org/10.1109/cdc49753.2023.10383375"},"language":"en","primary_location":{"id":"doi:10.1109/cdc49753.2023.10383375","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cdc49753.2023.10383375","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 62nd IEEE Conference on Decision and Control (CDC)","raw_type":"proceedings-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/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":["Tsinghua University,Department of Mechanical Engineering,Beijing,China,100084"],"affiliations":[{"raw_affiliation_string":"Tsinghua University,Department of Mechanical Engineering,Beijing,China,100084","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":["Tsinghua University,Department of Mechanical Engineering,Beijing,China,100084"],"affiliations":[{"raw_affiliation_string":"Tsinghua University,Department of Mechanical Engineering,Beijing,China,100084","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","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, Tsinghua University,Beijing,China,100084"],"affiliations":[{"raw_affiliation_string":"School of Vehicle and Mobility, Tsinghua University,Beijing,China,100084","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102566877","display_name":"Jia Cheng","orcid":null},"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":"Jia Cheng","raw_affiliation_strings":["Tsinghua University,Department of Mechanical Engineering,Beijing,China,100084"],"affiliations":[{"raw_affiliation_string":"Tsinghua University,Department of Mechanical Engineering,Beijing,China,100084","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"last","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":["Tsinghua University,Department of Mechanical Engineering,Beijing,China,100084"],"affiliations":[{"raw_affiliation_string":"Tsinghua University,Department of Mechanical Engineering,Beijing,China,100084","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":0.5237,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.73182835,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"334","last_page":"341"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.9984999895095825,"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":0.9984999895095825,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9731000065803528,"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":0.97079998254776,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.8479695916175842},{"id":"https://openalex.org/keywords/adversarial-system","display_name":"Adversarial system","score":0.709761917591095},{"id":"https://openalex.org/keywords/reachability","display_name":"Reachability","score":0.6910823583602905},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6552261114120483},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.5649276971817017},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.5415336489677429},{"id":"https://openalex.org/keywords/bellman-equation","display_name":"Bellman equation","score":0.4922267496585846},{"id":"https://openalex.org/keywords/invariant","display_name":"Invariant (physics)","score":0.4860575497150421},{"id":"https://openalex.org/keywords/adversary","display_name":"Adversary","score":0.43317461013793945},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.42887449264526367},{"id":"https://openalex.org/keywords/robust-control","display_name":"Robust control","score":0.4214578866958618},{"id":"https://openalex.org/keywords/monotonic-function","display_name":"Monotonic function","score":0.41386252641677856},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3530135750770569},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.2662423253059387},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.2043101191520691},{"id":"https://openalex.org/keywords/control-system","display_name":"Control system","score":0.18599292635917664},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.09945422410964966},{"id":"https://openalex.org/keywords/computer-security","display_name":"Computer security","score":0.09232598543167114}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.8479695916175842},{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.709761917591095},{"id":"https://openalex.org/C136643341","wikidata":"https://www.wikidata.org/wiki/Q1361526","display_name":"Reachability","level":2,"score":0.6910823583602905},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6552261114120483},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.5649276971817017},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.5415336489677429},{"id":"https://openalex.org/C14646407","wikidata":"https://www.wikidata.org/wiki/Q1430750","display_name":"Bellman equation","level":2,"score":0.4922267496585846},{"id":"https://openalex.org/C190470478","wikidata":"https://www.wikidata.org/wiki/Q2370229","display_name":"Invariant (physics)","level":2,"score":0.4860575497150421},{"id":"https://openalex.org/C41065033","wikidata":"https://www.wikidata.org/wiki/Q2825412","display_name":"Adversary","level":2,"score":0.43317461013793945},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.42887449264526367},{"id":"https://openalex.org/C31531917","wikidata":"https://www.wikidata.org/wiki/Q915157","display_name":"Robust control","level":3,"score":0.4214578866958618},{"id":"https://openalex.org/C72169020","wikidata":"https://www.wikidata.org/wiki/Q194404","display_name":"Monotonic function","level":2,"score":0.41386252641677856},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3530135750770569},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2662423253059387},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2043101191520691},{"id":"https://openalex.org/C17500928","wikidata":"https://www.wikidata.org/wiki/Q959968","display_name":"Control system","level":2,"score":0.18599292635917664},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.09945422410964966},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.09232598543167114},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","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/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"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/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/cdc49753.2023.10383375","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cdc49753.2023.10383375","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 62nd IEEE Conference on Decision and Control (CDC)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Peace, Justice and strong institutions","score":0.8100000023841858,"id":"https://metadata.un.org/sdg/16"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":39,"referenced_works":["https://openalex.org/W1678380195","https://openalex.org/W1845972764","https://openalex.org/W1880805667","https://openalex.org/W2028560736","https://openalex.org/W2106816694","https://openalex.org/W2158782408","https://openalex.org/W2171505650","https://openalex.org/W2498694880","https://openalex.org/W2560504659","https://openalex.org/W2766447205","https://openalex.org/W2964040381","https://openalex.org/W2968547875","https://openalex.org/W2968945909","https://openalex.org/W3088310808","https://openalex.org/W3120778962","https://openalex.org/W3127352841","https://openalex.org/W3127561923","https://openalex.org/W3135454970","https://openalex.org/W3162902207","https://openalex.org/W3195968524","https://openalex.org/W4206497039","https://openalex.org/W4224216677","https://openalex.org/W4226245271","https://openalex.org/W4293370597","https://openalex.org/W4317553716","https://openalex.org/W4362650413","https://openalex.org/W4385767976","https://openalex.org/W6638036514","https://openalex.org/W6639175102","https://openalex.org/W6687063787","https://openalex.org/W6701090783","https://openalex.org/W6747473740","https://openalex.org/W6751725685","https://openalex.org/W6764969207","https://openalex.org/W6771280675","https://openalex.org/W6774126978","https://openalex.org/W6780587392","https://openalex.org/W6804655846","https://openalex.org/W6838566501"],"related_works":["https://openalex.org/W2136512912","https://openalex.org/W2127267268","https://openalex.org/W2156446763","https://openalex.org/W2067910792","https://openalex.org/W2150194458","https://openalex.org/W2143461633","https://openalex.org/W2889071233","https://openalex.org/W4321471459","https://openalex.org/W4221156924","https://openalex.org/W2380814829"],"abstract_inverted_index":{"Safety":[0],"is":[1,80,89],"a":[2,45,59,71,92,155,183],"primary":[3],"concern":[4],"when":[5],"applying":[6],"reinforcement":[7,23,48,157],"learning":[8,24,49,158,182],"to":[9,63,90,114,124,139],"real-world":[10],"control":[11,111,198],"tasks,":[12],"especially":[13],"in":[14,36,102,106,234],"the":[15,66,74,98,108,118,132,140,149,163,202,219,231,235,238],"presence":[16],"of":[17,73,237],"external":[18,29,121],"disturbances.":[19,54,194],"However,":[20],"existing":[21],"safe":[22,47,75],"algorithms":[25,217],"rarely":[26],"account":[27],"for":[28,65,83,170],"disturbances,":[30,213],"limiting":[31],"their":[32],"applicability":[33],"and":[34,117,167,179],"robustness":[35],"practice.":[37],"To":[38],"address":[39],"this":[40,42,56,146],"challenge,":[41],"paper":[43,57,129,147],"proposes":[44],"robust":[46,67,142,164],"framework":[50],"that":[51,131,160,185,201],"tackles":[52,176],"worst-case":[53,193,211],"First,":[55],"presents":[58],"policy":[60,134,151,172,184],"iteration":[61,135,152],"scheme":[62,153],"solve":[64],"invariant":[68,143,165],"set,":[69,76],"i.e.,":[70,181],"subset":[72],"where":[77],"persistent":[78],"safety":[79,99,116,191,220],"only":[81],"possible":[82],"states":[84],"within.":[85],"The":[86],"key":[87],"idea":[88],"establish":[91],"two-player":[93],"zero-sum":[94],"game":[95],"by":[96],"leveraging":[97],"value":[100],"function":[101],"Hamilton-Jacobi":[103],"reachability":[104],"analysis,":[105],"which":[107],"protagonist":[109],"(i.e.,":[110,120],"inputs)":[112],"aims":[113],"maintain":[115],"adversary":[119],"disturbances)":[122],"tries":[123],"break":[125],"down":[126],"safety.":[127],"This":[128,174],"proves":[130],"proposed":[133,150,203,224],"algorithm":[136,159,175],"converges":[137],"monotonically":[138],"maximal":[141],"set.":[144],"Second,":[145],"integrates":[148],"into":[154],"constrained":[156,171],"simultaneously":[161],"synthesizes":[162],"set":[166],"uses":[168],"it":[169],"optimization.":[173],"both":[177],"optimality":[178],"safety,":[180],"attains":[186,227],"high":[187],"rewards":[188],"while":[189,214],"maintaining":[190],"under":[192],"Experiments":[195],"on":[196],"classic":[197],"tasks":[199],"show":[200],"method":[204,225],"achieves":[205],"zero":[206],"constraint":[207],"violation":[208],"with":[209],"learned":[210],"adversarial":[212],"other":[215],"baseline":[216],"violate":[218],"constraints":[221],"substantially.":[222],"Our":[223],"also":[226],"comparable":[228],"performance":[229],"as":[230],"baselines":[232],"even":[233],"absence":[236],"adversary.":[239]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":1}],"updated_date":"2026-01-13T01:12:25.745995","created_date":"2025-10-10T00:00:00"}