{"id":"https://openalex.org/W3160598148","doi":"https://doi.org/10.1109/jsait.2021.3081108","title":"Cautious Reinforcement Learning via Distributional Risk in the Dual Domain","display_name":"Cautious Reinforcement Learning via Distributional Risk in the Dual Domain","publication_year":2021,"publication_date":"2021-05-17","ids":{"openalex":"https://openalex.org/W3160598148","doi":"https://doi.org/10.1109/jsait.2021.3081108","mag":"3160598148"},"language":"en","primary_location":{"id":"doi:10.1109/jsait.2021.3081108","is_oa":false,"landing_page_url":"https://doi.org/10.1109/jsait.2021.3081108","pdf_url":null,"source":{"id":"https://openalex.org/S4210211895","display_name":"IEEE Journal on Selected Areas in Information Theory","issn_l":"2641-8770","issn":["2641-8770"],"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 Journal on Selected Areas in Information Theory","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/A5101898224","display_name":"Junyu Zhang","orcid":"https://orcid.org/0000-0003-2194-9664"},"institutions":[{"id":"https://openalex.org/I20089843","display_name":"Princeton University","ror":"https://ror.org/00hx57361","country_code":"US","type":"education","lineage":["https://openalex.org/I20089843"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Junyu Zhang","raw_affiliation_strings":["Princeton University, Princeton, NJ, USA","Dept. of Operations Res. & Financial Eng., Princeton Univ. Princeton, NJ, USA"],"affiliations":[{"raw_affiliation_string":"Princeton University, Princeton, NJ, USA","institution_ids":["https://openalex.org/I20089843"]},{"raw_affiliation_string":"Dept. of Operations Res. & Financial Eng., Princeton Univ. Princeton, NJ, USA","institution_ids":["https://openalex.org/I20089843"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5039563144","display_name":"Amrit Singh Bedi","orcid":"https://orcid.org/0000-0002-8807-2695"},"institutions":[{"id":"https://openalex.org/I166416128","display_name":"DEVCOM Army Research Laboratory","ror":"https://ror.org/011hc8f90","country_code":"US","type":"government","lineage":["https://openalex.org/I1304082316","https://openalex.org/I1330347796","https://openalex.org/I166416128","https://openalex.org/I2802705668","https://openalex.org/I4210154437"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Amrit Singh Bedi","raw_affiliation_strings":["CISD, U.S. Army Research Laboratory, Adelphi, MD, USA","[CISD, U.S. Army Research Laboratory, Adelphi, MD, USA]"],"affiliations":[{"raw_affiliation_string":"CISD, U.S. Army Research Laboratory, Adelphi, MD, USA","institution_ids":["https://openalex.org/I166416128"]},{"raw_affiliation_string":"[CISD, U.S. Army Research Laboratory, Adelphi, MD, USA]","institution_ids":["https://openalex.org/I166416128"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100707460","display_name":"Mengdi Wang","orcid":"https://orcid.org/0000-0002-2101-9507"},"institutions":[{"id":"https://openalex.org/I20089843","display_name":"Princeton University","ror":"https://ror.org/00hx57361","country_code":"US","type":"education","lineage":["https://openalex.org/I20089843"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Mengdi Wang","raw_affiliation_strings":["Princeton University, Princeton, NJ, USA","Dept. of Operations Res. & Financial Eng., Princeton Univ. Princeton, NJ, USA"],"affiliations":[{"raw_affiliation_string":"Princeton University, Princeton, NJ, USA","institution_ids":["https://openalex.org/I20089843"]},{"raw_affiliation_string":"Dept. of Operations Res. & Financial Eng., Princeton Univ. Princeton, NJ, USA","institution_ids":["https://openalex.org/I20089843"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5025896653","display_name":"Alec Koppel","orcid":"https://orcid.org/0000-0003-2447-2873"},"institutions":[{"id":"https://openalex.org/I166416128","display_name":"DEVCOM Army Research Laboratory","ror":"https://ror.org/011hc8f90","country_code":"US","type":"government","lineage":["https://openalex.org/I1304082316","https://openalex.org/I1330347796","https://openalex.org/I166416128","https://openalex.org/I2802705668","https://openalex.org/I4210154437"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Alec Koppel","raw_affiliation_strings":["CISD, U.S. Army Research Laboratory, Adelphi, MD, USA","[CISD, U.S. Army Research Laboratory, Adelphi, MD, USA]"],"affiliations":[{"raw_affiliation_string":"CISD, U.S. Army Research Laboratory, Adelphi, MD, USA","institution_ids":["https://openalex.org/I166416128"]},{"raw_affiliation_string":"[CISD, U.S. Army Research Laboratory, Adelphi, MD, USA]","institution_ids":["https://openalex.org/I166416128"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5101898224"],"corresponding_institution_ids":["https://openalex.org/I20089843"],"apc_list":null,"apc_paid":null,"fwci":1.6316,"has_fulltext":false,"cited_by_count":13,"citation_normalized_percentile":{"value":0.86377196,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":98},"biblio":{"volume":"2","issue":"2","first_page":"611","last_page":"626"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.9962000250816345,"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":0.9962000250816345,"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/T11413","display_name":"Risk and Portfolio Optimization","score":0.9954000115394592,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T12101","display_name":"Advanced Bandit Algorithms Research","score":0.9818000197410583,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/risk-measure","display_name":"Risk measure","score":0.7120857238769531},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.6552985310554504},{"id":"https://openalex.org/keywords/markov-decision-process","display_name":"Markov decision process","score":0.6493033170700073},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.6126159429550171},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5475189089775085},{"id":"https://openalex.org/keywords/dynamic-risk-measure","display_name":"Dynamic risk measure","score":0.43444934487342834},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.43103542923927307},{"id":"https://openalex.org/keywords/measure","display_name":"Measure (data warehouse)","score":0.43000662326812744},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.4087323248386383},{"id":"https://openalex.org/keywords/markov-process","display_name":"Markov process","score":0.37421008944511414},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.18290975689888}],"concepts":[{"id":"https://openalex.org/C2781472820","wikidata":"https://www.wikidata.org/wiki/Q2154759","display_name":"Risk measure","level":3,"score":0.7120857238769531},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6552985310554504},{"id":"https://openalex.org/C106189395","wikidata":"https://www.wikidata.org/wiki/Q176789","display_name":"Markov decision process","level":3,"score":0.6493033170700073},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.6126159429550171},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5475189089775085},{"id":"https://openalex.org/C167548438","wikidata":"https://www.wikidata.org/wiki/Q5319012","display_name":"Dynamic risk measure","level":4,"score":0.43444934487342834},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.43103542923927307},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.43000662326812744},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.4087323248386383},{"id":"https://openalex.org/C159886148","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov process","level":2,"score":0.37421008944511414},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.18290975689888},{"id":"https://openalex.org/C106159729","wikidata":"https://www.wikidata.org/wiki/Q2294553","display_name":"Financial economics","level":1,"score":0.0},{"id":"https://openalex.org/C77088390","wikidata":"https://www.wikidata.org/wiki/Q8513","display_name":"Database","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},{"id":"https://openalex.org/C2780821815","wikidata":"https://www.wikidata.org/wiki/Q5340806","display_name":"Portfolio","level":2,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/jsait.2021.3081108","is_oa":false,"landing_page_url":"https://doi.org/10.1109/jsait.2021.3081108","pdf_url":null,"source":{"id":"https://openalex.org/S4210211895","display_name":"IEEE Journal on Selected Areas in Information Theory","issn_l":"2641-8770","issn":["2641-8770"],"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 Journal on Selected Areas in Information Theory","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Peace, Justice and strong institutions","score":0.7900000214576721,"id":"https://metadata.un.org/sdg/16"}],"awards":[{"id":"https://openalex.org/G7891574705","display_name":null,"funder_award_id":"W911NF-16-2-0008","funder_id":"https://openalex.org/F4320338295","funder_display_name":"Army Research Laboratory"}],"funders":[{"id":"https://openalex.org/F4320338295","display_name":"Army Research Laboratory","ror":"https://ror.org/011hc8f90"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":83,"referenced_works":["https://openalex.org/W51049863","https://openalex.org/W158549001","https://openalex.org/W1191599655","https://openalex.org/W1529558080","https://openalex.org/W1557287189","https://openalex.org/W1587828677","https://openalex.org/W1678814353","https://openalex.org/W1745373831","https://openalex.org/W1757796397","https://openalex.org/W1856231595","https://openalex.org/W1964535365","https://openalex.org/W1969147614","https://openalex.org/W1972190631","https://openalex.org/W1992586302","https://openalex.org/W2001009060","https://openalex.org/W2016384870","https://openalex.org/W2019291268","https://openalex.org/W2075090252","https://openalex.org/W2099506495","https://openalex.org/W2119567691","https://openalex.org/W2121863487","https://openalex.org/W2126282494","https://openalex.org/W2128521145","https://openalex.org/W2134293360","https://openalex.org/W2158191646","https://openalex.org/W2162849300","https://openalex.org/W2170923204","https://openalex.org/W2236244207","https://openalex.org/W2313791856","https://openalex.org/W2462780152","https://openalex.org/W2562316362","https://openalex.org/W2765892966","https://openalex.org/W2798543980","https://openalex.org/W2803308811","https://openalex.org/W2804589917","https://openalex.org/W2891086551","https://openalex.org/W2913434950","https://openalex.org/W2951520685","https://openalex.org/W2951966769","https://openalex.org/W2952647718","https://openalex.org/W2962803570","https://openalex.org/W2962878825","https://openalex.org/W2963082979","https://openalex.org/W2963423916","https://openalex.org/W2963438131","https://openalex.org/W2963856199","https://openalex.org/W2964108826","https://openalex.org/W2964340170","https://openalex.org/W2970036354","https://openalex.org/W2970927156","https://openalex.org/W2980326798","https://openalex.org/W2981915592","https://openalex.org/W2983617607","https://openalex.org/W2998059334","https://openalex.org/W3007273753","https://openalex.org/W3011338904","https://openalex.org/W3123298421","https://openalex.org/W3124407081","https://openalex.org/W3162902207","https://openalex.org/W3183820536","https://openalex.org/W4214717370","https://openalex.org/W4236670843","https://openalex.org/W4293545785","https://openalex.org/W4294562617","https://openalex.org/W4298857966","https://openalex.org/W4298876402","https://openalex.org/W6606376947","https://openalex.org/W6637429886","https://openalex.org/W6637967152","https://openalex.org/W6638914580","https://openalex.org/W6683300800","https://openalex.org/W6683569045","https://openalex.org/W6685029077","https://openalex.org/W6687063787","https://openalex.org/W6737893269","https://openalex.org/W6751535212","https://openalex.org/W6751629939","https://openalex.org/W6751857639","https://openalex.org/W6759052742","https://openalex.org/W6767112054","https://openalex.org/W6767486501","https://openalex.org/W6770329903","https://openalex.org/W6774146290"],"related_works":["https://openalex.org/W4286713570","https://openalex.org/W3213838085","https://openalex.org/W4286693834","https://openalex.org/W4313591620","https://openalex.org/W2390529924","https://openalex.org/W2197677645","https://openalex.org/W2998948027","https://openalex.org/W2937181779","https://openalex.org/W1985560493","https://openalex.org/W4226345396"],"abstract_inverted_index":{"We":[0,150],"study":[1],"the":[2,37,82,85,95,106,113,120,153,165,168,174,178,184,194,210,232,241],"estimation":[3],"of":[4,47,70,84,90,98,105,119,138,167,177,193,212,234],"risk-sensitive":[5,40],"policies":[6],"in":[7,129],"reinforcement":[8],"learning":[9],"(RL)":[10],"problems":[11],"defined":[12],"by":[13,32],"a":[14,67,77,99,103,116,135,190,238],"Markov":[15],"Decision":[16],"Process":[17],"(MDPs)":[18],"whose":[19],"state":[20,109],"and":[21,44,101,197,227,240],"action":[22],"spaces":[23],"are":[24,29,42,50],"countably":[25],"finite.":[26],"Prior":[27],"efforts":[28],"predominately":[30],"afflicted":[31],"computational":[33],"challenges":[34],"associated":[35],"with":[36,134],"fact":[38],"that":[39,141,152,206],"MDPs":[41],"time-inconsistent,":[43],"hence":[45],"modifications":[46],"Bellman's":[48],"equations":[49],"required,":[51],"which":[52,72],"often":[53],"do":[54],"not":[55],"permit":[56],"efficient":[57],"fixed":[58],"point":[59],"schemes.":[60],"To":[61],"ameliorate":[62],"this":[63,127,207],"issue,":[64],"we":[65,73,123,188],"propose":[66,124,189],"new":[68],"definition":[69],"risk,":[71],"call":[74],"caution,":[75],"as":[76,146,218],"penalty":[78],"function":[79,104,118],"added":[80],"to":[81,125,157,201,220,237],"dual":[83],"linear":[86],"programming":[87],"(LP)":[88],"formulation":[89],"tabular":[91],"RL.":[92],"Caution":[93],"quantifies":[94],"distributional":[96],"risk":[97,114,179,185],"policy,":[100],"is":[102,115,186],"policy's":[107],"long-term":[108],"occupancy":[110,121],"measure.":[111],"When":[112],"convex":[117,226],"measure,":[122],"solve":[126],"problem":[128],"an":[130],"online":[131],"model-free":[132],"manner":[133],"stochastic":[136],"variant":[137],"primal-dual":[139],"method":[140],"uses":[142],"Kullback-Lieber":[143],"(KL)":[144],"divergence":[145,236],"its":[147,199],"proximal":[148],"term.":[149],"establish":[151,198],"sample":[154],"complexity":[155],"required":[156],"attain":[158],"near-optimal":[159],"solutions":[160],"matches":[161],"tight":[162],"dependencies":[163],"on":[164,173],"cardinality":[166],"spaces,":[169],"but":[170],"also":[171],"depends":[172],"infinity":[175],"norm":[176],"measure's":[180],"gradient.":[181],"Moreover,":[182],"when":[183],"non-convex,":[187],"block-coordinate":[191],"augmentation":[192],"aforementioned":[195],"approach,":[196],"convergence":[200],"KKT":[202],"points.":[203],"Experiments":[204],"demonstrate":[205],"approach":[208],"improves":[209],"reliability":[211],"reward":[213],"accumulation":[214],"without":[215],"additional":[216],"computation":[217],"compared":[219],"risk-neutral":[221],"LP":[222],"solvers,":[223],"both":[224],"for":[225,231],"non-convex":[228],"risks,":[229],"respectively,":[230],"cases":[233],"KL":[235],"prior,":[239],"variance.":[242]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":3},{"year":2023,"cited_by_count":2},{"year":2022,"cited_by_count":5},{"year":2021,"cited_by_count":2}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}