{"id":"https://openalex.org/W4223902497","doi":"https://doi.org/10.1109/ciss53076.2022.9751158","title":"Doubly Pessimistic Algorithms for Strictly Safe Off-Policy Optimization","display_name":"Doubly Pessimistic Algorithms for Strictly Safe Off-Policy Optimization","publication_year":2022,"publication_date":"2022-03-09","ids":{"openalex":"https://openalex.org/W4223902497","doi":"https://doi.org/10.1109/ciss53076.2022.9751158"},"language":"en","primary_location":{"is_oa":false,"landing_page_url":"https://doi.org/10.1109/ciss53076.2022.9751158","pdf_url":null,"source":null,"license":null,"version":null,"is_accepted":false,"is_published":false},"type":"article","type_crossref":"proceedings-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/A5011093549","display_name":"Sanae Amani","orcid":null},"institutions":[{"id":"https://openalex.org/I2799798094","display_name":"UCLA Health","ror":"https://ror.org/01d88se56","country_code":"US","type":"healthcare","lineage":["https://openalex.org/I2799798094"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Sanae Amani","raw_affiliation_string":"UCLA","raw_affiliation_strings":["UCLA"]},{"author_position":"last","author":{"id":"https://openalex.org/A5072096775","display_name":"Lin F. Yang","orcid":"https://orcid.org/0000-0001-9056-0500"},"institutions":[{"id":"https://openalex.org/I2799798094","display_name":"UCLA Health","ror":"https://ror.org/01d88se56","country_code":"US","type":"healthcare","lineage":["https://openalex.org/I2799798094"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Lin F. Yang","raw_affiliation_string":"UCLA","raw_affiliation_strings":["UCLA"]}],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"has_fulltext":false,"cited_by_count":0,"cited_by_percentile_year":{"min":0,"max":69},"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"primary_topic":{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning Algorithms","score":0.9977,"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 Algorithms","score":0.9977,"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 Deep Learning Models","score":0.9972,"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/T12535","display_name":"Learning with Noisy Labels in Machine Learning","score":0.9848,"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":[{"keyword":"pessimistic algorithms","score":0.629},{"keyword":"optimization","score":0.4138},{"keyword":"off-policy","score":0.25}],"concepts":[{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5990931},{"id":"https://openalex.org/C2776036281","wikidata":"https://www.wikidata.org/wiki/Q48769818","display_name":"Constraint (computer-aided design)","level":2,"score":0.5717682},{"id":"https://openalex.org/C42747912","wikidata":"https://www.wikidata.org/wiki/Q1048447","display_name":"Multiplicative function","level":2,"score":0.56481874},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5623873},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.49902797},{"id":"https://openalex.org/C106189395","wikidata":"https://www.wikidata.org/wiki/Q176789","display_name":"Markov decision process","level":3,"score":0.49690774},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.48299205},{"id":"https://openalex.org/C14646407","wikidata":"https://www.wikidata.org/wiki/Q1430750","display_name":"Bellman equation","level":2,"score":0.42224243},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.38236147},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3340092},{"id":"https://openalex.org/C159886148","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov process","level":2,"score":0.29627198},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.10300967},{"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/C151730666","wikidata":"https://www.wikidata.org/wiki/Q7205","display_name":"Paleontology","level":1,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"is_oa":false,"landing_page_url":"https://doi.org/10.1109/ciss53076.2022.9751158","pdf_url":null,"source":null,"license":null,"version":null,"is_accepted":false,"is_published":false}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Peace, justice, and strong institutions","id":"https://metadata.un.org/sdg/16","score":0.77}],"grants":[],"referenced_works_count":2,"referenced_works":["https://openalex.org/W192920577","https://openalex.org/W3150718622"],"related_works":["https://openalex.org/W2152670157","https://openalex.org/W4308702637","https://openalex.org/W2386410636","https://openalex.org/W176737593","https://openalex.org/W2156021013","https://openalex.org/W2903299703","https://openalex.org/W2016648086","https://openalex.org/W2808418668","https://openalex.org/W4385342861","https://openalex.org/W2062076630"],"ngrams_url":"https://api.openalex.org/works/W4223902497/ngrams","abstract_inverted_index":{"We":[0,38,70,146,199],"study":[1],"offline":[2,172],"reinforcement":[3],"learning":[4],"(RL)":[5],"in":[6,75,88],"the":[7,23,35,44,62,76,117,121,127,137,141,190],"presence":[8],"of":[9,52,78,100,120,140,169,217],"safety":[10,36,45,191],"requirements:":[11],"from":[12,111],"a":[13,16,67,89,97,108,132],"dataset":[14,122,158],"collected":[15],"priori":[17],"and":[18,54,103,163],"without":[19],"direct":[20],"access":[21],"to":[22,33,61,151,213],"true":[24],"environment,":[25],"learn":[26],"an":[27,48,73,176],"optimal":[28],"policy":[29,63,110,143],"that":[30,86,112,168,204],"is":[31,220],"guaranteed":[32],"respect":[34,60],"constraints.":[37],"address":[39],"this":[40,218],"problem":[41],"by":[42],"modeling":[43],"requirement":[46],"as":[47,214],"unknown":[49],"cost":[50],"function":[51],"states":[53],"actions,":[55],"whose":[56],"expected":[57],"value":[58],"with":[59,154,175],"must":[64],"fall":[65],"below":[66],"certain":[68],"threshold.":[69],"then":[71,147],"present":[72,201],"algorithm":[74],"context":[77],"finite-horizon":[79],"Markov":[80],"decision":[81],"processes":[82],"(MDPs),":[83],"termed":[84],"Safe-DPVI":[85,144],"performs":[87],"doubly":[90],"pessimistic":[91],"manner":[92],"when":[93,105],"1)":[94],"it":[95,106],"constructs":[96],"conservative":[98,113],"set":[99],"safe":[101,142],"policies;":[102],"2)":[104],"selects":[107],"good":[109],"set.":[114],"Without":[115],"assuming":[116],"sufficient":[118],"coverage":[119],"or":[123],"any":[124],"structure":[125],"for":[126],"underlying":[128],"MDPs,":[129],"we":[130],"establish":[131],"data-dependent":[133,162],"upper":[134],"bound":[135],"on":[136,157],"suboptimality":[138],"gap":[139],"returns.":[145],"specialize":[148],"our":[149,206],"results":[150],"linear":[152],"MDPs":[153],"appropriate":[155],"assumptions":[156],"being":[159],"well-explored.":[160],"Both":[161],"specialized":[164],"bounds":[165],"nearly":[166],"match":[167],"state-of-the-art":[170],"unsafe":[171],"RL":[173],"algorithms,":[174],"additional":[177],"multiplicative":[178],"factor":[179],"$\\frac{\\Sigma_{h=1}^{H}\\alpha_{h}}{H}$":[182],",":[183],"where":[184],"\u03b1":[185],"h":[188],"characterizes":[189],"constraint":[192],"at":[193],"time-step":[194],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">$h$":[197],".":[198],"further":[200],"numerical":[202],"simulations":[203],"corroborate":[205],"theoretical":[207],"findings.":[208],"A":[209],"full":[210],"version":[211],"referred":[212],"technical":[215],"report":[216],"paper":[219],"accessible":[221],"at:":[222],"https://offline-rl-neurips.github.io/2021/pdf/21.pdf":[223]},"cited_by_api_url":"https://api.openalex.org/works?filter=cites:W4223902497","counts_by_year":[],"updated_date":"2024-02-21T07:23:08.773395","created_date":"2022-04-19"}