{"id":"https://openalex.org/W3011080757","doi":"https://doi.org/10.1109/cdc40024.2019.9030265","title":"Convergence and Iteration Complexity of Policy Gradient Method for Infinite-horizon Reinforcement Learning","display_name":"Convergence and Iteration Complexity of Policy Gradient Method for Infinite-horizon Reinforcement Learning","publication_year":2019,"publication_date":"2019-12-01","ids":{"openalex":"https://openalex.org/W3011080757","doi":"https://doi.org/10.1109/cdc40024.2019.9030265","mag":"3011080757"},"language":"en","primary_location":{"is_oa":false,"landing_page_url":"https://doi.org/10.1109/cdc40024.2019.9030265","pdf_url":null,"source":null,"license":null,"license_id":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/A5047410441","display_name":"Kaiqing Zhang","orcid":"https://orcid.org/0000-0002-7446-7581"},"institutions":[{"id":"https://openalex.org/I157725225","display_name":"University of Illinois Urbana-Champaign","ror":"https://ror.org/047426m28","country_code":"US","type":"education","lineage":["https://openalex.org/I157725225"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kaiqing Zhang","raw_affiliation_strings":["University of Illinois at Urbana-Champaign - Coordinated Science Laboratory"],"affiliations":[{"raw_affiliation_string":"University of Illinois at Urbana-Champaign - Coordinated Science Laboratory","institution_ids":["https://openalex.org/I157725225"]}]},{"author_position":"middle","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/I4210088792","https://openalex.org/I4210154437"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Alec Koppel","raw_affiliation_strings":["U.S. Army Research Laboratory,Computational and Information Sciences Directorate,Adelphi,MD,USA"],"affiliations":[{"raw_affiliation_string":"U.S. Army Research Laboratory,Computational and Information Sciences Directorate,Adelphi,MD,USA","institution_ids":["https://openalex.org/I166416128"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100644862","display_name":"Hao Zhu","orcid":"https://orcid.org/0000-0002-1554-5332"},"institutions":[{"id":"https://openalex.org/I86519309","display_name":"The University of Texas at Austin","ror":"https://ror.org/00hj54h04","country_code":"US","type":"education","lineage":["https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hao Zhu","raw_affiliation_strings":["University of Texas at Austin,Dept. of Electrical and Computer Engineering"],"affiliations":[{"raw_affiliation_string":"University of Texas at Austin,Dept. of Electrical and Computer Engineering","institution_ids":["https://openalex.org/I86519309"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5019604570","display_name":"Tamer Ba\u015far","orcid":"https://orcid.org/0000-0003-4406-7875"},"institutions":[{"id":"https://openalex.org/I157725225","display_name":"University of Illinois Urbana-Champaign","ror":"https://ror.org/047426m28","country_code":"US","type":"education","lineage":["https://openalex.org/I157725225"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Tamer Basar","raw_affiliation_strings":["University of Illinois at Urbana-Champaign - Coordinated Science Laboratory"],"affiliations":[{"raw_affiliation_string":"University of Illinois at Urbana-Champaign - Coordinated Science Laboratory","institution_ids":["https://openalex.org/I157725225"]}]}],"institution_assertions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.25,"has_fulltext":false,"cited_by_count":7,"citation_normalized_percentile":{"value":0.589695,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":83,"max":84},"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":0.9991,"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":0.9991,"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/T11612","display_name":"Optimization Methods in Machine Learning","score":0.9981,"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/T12101","display_name":"Optimization of Multi-Armed Bandit Problems","score":0.997,"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/reinforcement-learning","display_name":"Reinforcement Learning","score":0.586312},{"id":"https://openalex.org/keywords/policy-gradient","display_name":"Policy Gradient","score":0.578865},{"id":"https://openalex.org/keywords/approximation-algorithms","display_name":"Approximation Algorithms","score":0.527732},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"Stochastic Gradient Descent","score":0.517808},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.512891}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.78258455},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.61892796},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.54488087},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.5233908},{"id":"https://openalex.org/C57869625","wikidata":"https://www.wikidata.org/wiki/Q1783502","display_name":"Rate of convergence","level":3,"score":0.5132675},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.45939657},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.45471144},{"id":"https://openalex.org/C159176650","wikidata":"https://www.wikidata.org/wiki/Q43261","display_name":"Horizon","level":2,"score":0.42052174},{"id":"https://openalex.org/C115680565","wikidata":"https://www.wikidata.org/wiki/Q5977448","display_name":"Gradient method","level":2,"score":0.41865373},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.39192834},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.26635414},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.08012983},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"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/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.0},{"id":"https://openalex.org/C127162648","wikidata":"https://www.wikidata.org/wiki/Q16858953","display_name":"Channel (broadcasting)","level":2,"score":0.0},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"is_oa":false,"landing_page_url":"https://doi.org/10.1109/cdc40024.2019.9030265","pdf_url":null,"source":null,"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Peace, justice, and strong institutions","score":0.45,"id":"https://metadata.un.org/sdg/16"}],"grants":[],"datasets":[],"versions":[],"referenced_works_count":36,"referenced_works":["https://openalex.org/W1577337550","https://openalex.org/W1697075315","https://openalex.org/W1771410628","https://openalex.org/W2046859786","https://openalex.org/W2082261506","https://openalex.org/W2094387729","https://openalex.org/W2098432798","https://openalex.org/W2113501460","https://openalex.org/W2119717200","https://openalex.org/W2121863487","https://openalex.org/W2136602922","https://openalex.org/W2155027007","https://openalex.org/W2592651140","https://openalex.org/W2766447205","https://openalex.org/W2783637925","https://openalex.org/W2788115019","https://openalex.org/W2901573274","https://openalex.org/W2904058351","https://openalex.org/W2912747791","https://openalex.org/W2913326990","https://openalex.org/W2947577977","https://openalex.org/W2950882283","https://openalex.org/W2952164720","https://openalex.org/W2963470657","https://openalex.org/W2963672698","https://openalex.org/W2963864421","https://openalex.org/W2964043796","https://openalex.org/W3109546547","https://openalex.org/W3115592686","https://openalex.org/W4232753567","https://openalex.org/W4247165901","https://openalex.org/W4289760659","https://openalex.org/W4293775970","https://openalex.org/W4302570325","https://openalex.org/W4302578309","https://openalex.org/W594357522"],"related_works":["https://openalex.org/W937190428","https://openalex.org/W4378552426","https://openalex.org/W4308234978","https://openalex.org/W4298544744","https://openalex.org/W3114199884","https://openalex.org/W3098788150","https://openalex.org/W3011663284","https://openalex.org/W2524032428","https://openalex.org/W1972717324","https://openalex.org/W1666485427"],"abstract_inverted_index":{"We":[0],"focus":[1],"on":[2,200],"policy":[3,34,120,128,172],"search":[4,129],"in":[5,155,174,188],"reinforcement":[6],"learning":[7],"problems":[8,41],"over":[9],"continuous":[10,43],"spaces,":[11],"where":[12],"the":[13,24,115,119,146,152,156,175,189,201,205],"value":[14],"is":[15,23,54,63,73],"defined":[16],"by":[17,28],"infinite-horizon":[18,176],"discounted":[19],"reward":[20],"accumulation.":[21],"This":[22],"canonical":[25],"setting":[26],"proposed":[27],"Bellman":[29],"[3].":[30],"Policy":[31],"search,":[32],"specifically,":[33],"gradient":[35,173,186],"(PG)":[36],"method,":[37],"scales":[38],"gracefully":[39],"to":[40,56,85,92,151,184],"with":[42],"spaces":[44],"and":[45,59,80,194],"allows":[46],"for":[47,114,192],"deep":[48],"network":[49],"parametrizations;":[50],"however,":[51],"experimentally":[52],"it":[53,180],"known":[55],"be":[57,88],"volatile":[58],"its":[60],"finite-time":[61],"behavior":[62],"not":[64],"well":[65],"understood.":[66],"A":[67],"major":[68],"source":[69],"of":[70,105,118,148,171,207],"this":[71,98],"gap":[72],"that":[74,108,179],"unbiased":[75,127],"ascent":[76],"directions":[77],"are":[78],"elusive,":[79],"hence":[81],"only":[82],"asymptotic":[83,149],"convergence":[84,135,150,169],"stationarity":[86],"can":[87],"shown":[89],"via":[90],"links":[91],"ordinary":[93],"differential":[94],"equations":[95],"[4].":[96],"In":[97],"work,":[99],"we":[100,123,132,143,164],"propose":[101],"a":[102,110,138],"new":[103],"variant":[104],"PG":[106],"methods":[107],"uses":[109],"random":[111],"rollout":[112],"horizon":[113],"Monte-Carlo":[116],"estimation":[117],"gradient,":[121],"which":[122],"establish":[124],"yields":[125],"an":[126,159],"direction.":[130],"Furthermore,":[131],"conduct":[133],"global":[134],"analysis":[136],"from":[137],"nonconvex":[139,190],"optimization":[140],"perspective:":[141],"(i)":[142],"first":[144],"recover":[145],"results":[147],"stationary-point":[153],"policies":[154],"literature":[157],"through":[158],"alternative":[160],"supermartingale":[161],"argument;":[162],"(ii)":[163],"provide":[165],"iteration":[166],"complexity,":[167],"i.e.,":[168],"rate,":[170],"setting,":[177],"showing":[178],"exhibits":[181],"comparable":[182],"rates":[183],"stochastic":[185],"method":[187],"regime":[191],"diminishing":[193],"constant":[195],"stepsize":[196],"rules.":[197],"Numerical":[198],"experiments":[199],"inverted":[202],"pendulum":[203],"demonstrate":[204],"validity":[206],"our":[208],"results.":[209]},"cited_by_api_url":"https://api.openalex.org/works?filter=cites:W3011080757","counts_by_year":[{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":1},{"year":2022,"cited_by_count":2},{"year":2021,"cited_by_count":3}],"updated_date":"2024-10-09T10:54:37.014870","created_date":"2020-03-23"}
