{"id":"https://openalex.org/W2914920107","doi":"https://doi.org/10.48550/arxiv.1901.11275","title":"A Theory of Regularized Markov Decision Processes","display_name":"A Theory of Regularized Markov Decision Processes","publication_year":2019,"publication_date":"2019-01-31","ids":{"openalex":"https://openalex.org/W2914920107","doi":"https://doi.org/10.48550/arxiv.1901.11275","mag":"2914920107"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:1901.11275","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/1901.11275","pdf_url":"https://arxiv.org/pdf/1901.11275","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/1901.11275","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5110482875","display_name":"Matthieu Geist","orcid":"https://orcid.org/0009-0002-7888-8233","observed_orcids":["https://orcid.org/0009-0002-7888-8233"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Geist, Matthieu","raw_affiliation_strings":["Google Brain, Paris (8 Rue de Londres-15-15 Bis Rue de Clichy 75009 Paris France\r\n - France)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google Brain, Paris (8 Rue de Londres-15-15 Bis Rue de Clichy 75009 Paris France\r\n - France)","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5032281855","display_name":"Bruno Scherrer","orcid":"https://orcid.org/0000-0001-6097-4489","observed_orcids":["https://orcid.org/0000-0001-6097-4489"]},"institutions":[{"id":"https://openalex.org/I3017839442","display_name":"Institut \u00c9lie Cartan de Lorraine","ror":"https://ror.org/04rvw8791","country_code":"FR","type":"facility","lineage":["https://openalex.org/I1294671590","https://openalex.org/I1326498283","https://openalex.org/I3017839442","https://openalex.org/I4210141950","https://openalex.org/I90183372"]},{"id":"https://openalex.org/I90183372","display_name":"Universit\u00e9 de Lorraine","ror":"https://ror.org/04vfs2w97","country_code":"FR","type":"education","lineage":["https://openalex.org/I90183372"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Scherrer, Bruno","raw_affiliation_strings":["BIGS - Biology, genetics and statistics (France)","IECL - Institut \u00c9lie Cartan de Lorraine (Universit\u00e9 de Lorraine, Boulevard des Aiguillettes BP 70239 54506 Vandoeuvre-les-Nancy Cedex\r\nIle du Saulcy - 57 045 Metz Cedex 01 - France)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"BIGS - Biology, genetics and statistics (France)","institution_ids":[]},{"raw_affiliation_string":"IECL - Institut \u00c9lie Cartan de Lorraine (Universit\u00e9 de Lorraine, Boulevard des Aiguillettes BP 70239 54506 Vandoeuvre-les-Nancy Cedex\r\nIle du Saulcy - 57 045 Metz Cedex 01 - France)","institution_ids":["https://openalex.org/I3017839442","https://openalex.org/I90183372"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5065100569","display_name":"Olivier Pietquin","orcid":"https://orcid.org/0000-0002-5386-465X","observed_orcids":["https://orcid.org/0000-0002-5386-465X"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pietquin, Olivier","raw_affiliation_strings":["Google Brain, Paris (8 Rue de Londres-15-15 Bis Rue de Clichy 75009 Paris France\r\n - France)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google Brain, Paris (8 Rue de Londres-15-15 Bis Rue de Clichy 75009 Paris France\r\n - France)","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":146,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.9990000128746033,"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":0.9990000128746033,"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":0.0008999999845400453,"subfield":{"id":"https://openalex.org/subfields/2207","display_name":"Control and Systems Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12101","display_name":"Advanced Bandit Algorithms Research","score":0.0,"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/regularization","display_name":"regularization","score":0.4429999887943268},{"id":"https://openalex.org/keywords/entropy-regularization","display_name":"entropy regularization","score":0.7269999980926514},{"id":"https://openalex.org/keywords/kullback-leibler-divergence","display_name":"Kullback-Leibler divergence","score":0.9869999885559082},{"id":"https://openalex.org/keywords/legendre-fenchel-transform","display_name":"Legendre-Fenchel transform","score":0.9990000128746033},{"id":"https://openalex.org/keywords/error-propagation-analysis","display_name":"error propagation analysis","score":0.746999979019165},{"id":"https://openalex.org/keywords/trust-region-policy-optimization","display_name":"trust region policy optimization","score":0.9020000100135803},{"id":"https://openalex.org/keywords/soft-q-learning","display_name":"soft Q-learning","score":0.9950000047683716},{"id":"https://openalex.org/keywords/mirror-descent","display_name":"mirror descent","score":0.8809999823570251}],"concepts":[{"id":"https://openalex.org/C106189395","wikidata":"https://www.wikidata.org/wiki/Q176789","display_name":"Markov decision process","level":3,"score":0.827864408493042},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6679582595825195},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.6302245259284973},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.537390410900116},{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.5295522809028625},{"id":"https://openalex.org/C37404715","wikidata":"https://www.wikidata.org/wiki/Q380679","display_name":"Dynamic programming","level":2,"score":0.5042279958724976},{"id":"https://openalex.org/C157972887","wikidata":"https://www.wikidata.org/wiki/Q463359","display_name":"Convex optimization","level":3,"score":0.49503985047340393},{"id":"https://openalex.org/C98763669","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov chain","level":2,"score":0.42653408646583557},{"id":"https://openalex.org/C145446738","wikidata":"https://www.wikidata.org/wiki/Q319913","display_name":"Convex function","level":3,"score":0.42156416177749634},{"id":"https://openalex.org/C159886148","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov process","level":2,"score":0.41704633831977844},{"id":"https://openalex.org/C17020691","wikidata":"https://www.wikidata.org/wiki/Q139677","display_name":"Operator (biology)","level":5,"score":0.41253340244293213},{"id":"https://openalex.org/C106301342","wikidata":"https://www.wikidata.org/wiki/Q4117933","display_name":"Entropy (arrow of time)","level":2,"score":0.41229718923568726},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.4078810214996338},{"id":"https://openalex.org/C112680207","wikidata":"https://www.wikidata.org/wiki/Q714886","display_name":"Regular polygon","level":2,"score":0.3976987600326538},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2745240330696106},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.1352030634880066},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","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/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"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/C158448853","wikidata":"https://www.wikidata.org/wiki/Q425218","display_name":"Repressor","level":4,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C86339819","wikidata":"https://www.wikidata.org/wiki/Q407384","display_name":"Transcription factor","level":3,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0}],"mesh":[],"locations_count":3,"locations":[{"id":"pmh:oai:arXiv.org:1901.11275","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/1901.11275","pdf_url":"https://arxiv.org/pdf/1901.11275","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"pmh:oai:HAL:hal-02273741v1","doi":null,"is_oa":false,"landing_page_url":"https://inria.hal.science/hal-02273741","pdf_url":null,"source":{"id":"https://openalex.org/S4306402512","display_name":"HAL","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I4405252976","host_organization_name":"Centre pour la Communication Scientifique Directe","host_organization_lineage":["https://openalex.org/I4405252976"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"ICML 2019 - Thirty-sixth International Conference on Machine Learning, Jun 2019, Long Island, United States","raw_type":"info:eu-repo/semantics/conferenceObject"},{"id":"doi:10.48550/arxiv.1901.11275","doi":"https://doi.org/10.48550/arxiv.1901.11275","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.1901.11275","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:1901.11275","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/1901.11275","pdf_url":"https://arxiv.org/pdf/1901.11275","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W2914920107.pdf","grobid_xml":"https://content.openalex.org/works/W2914920107.grobid-xml"},"referenced_works_count":40,"referenced_works":["https://openalex.org/W1499669280","https://openalex.org/W1503425191","https://openalex.org/W1576452626","https://openalex.org/W1771410628","https://openalex.org/W1777239053","https://openalex.org/W1788877992","https://openalex.org/W1889629917","https://openalex.org/W2016384870","https://openalex.org/W2020609518","https://openalex.org/W2098774185","https://openalex.org/W2119717200","https://openalex.org/W2144846366","https://openalex.org/W2155027007","https://openalex.org/W2167732364","https://openalex.org/W2169713291","https://openalex.org/W2253795368","https://openalex.org/W2260756217","https://openalex.org/W2290104316","https://openalex.org/W2411690432","https://openalex.org/W2593044849","https://openalex.org/W2594103415","https://openalex.org/W2609650878","https://openalex.org/W2619268125","https://openalex.org/W2755631392","https://openalex.org/W2766610320","https://openalex.org/W2776810666","https://openalex.org/W2781726626","https://openalex.org/W2785738552","https://openalex.org/W2787063232","https://openalex.org/W2788781499","https://openalex.org/W2799151646","https://openalex.org/W2806985155","https://openalex.org/W2902286283","https://openalex.org/W2904246096","https://openalex.org/W2949608212","https://openalex.org/W2950422748","https://openalex.org/W2962777832","https://openalex.org/W2963169817","https://openalex.org/W2963267001","https://openalex.org/W6990719774"],"related_works":["https://openalex.org/W2619268125","https://openalex.org/W4381739798","https://openalex.org/W2973365583","https://openalex.org/W4391985038","https://openalex.org/W4387596101","https://openalex.org/W7130606674","https://openalex.org/W4320812399","https://openalex.org/W4324325663","https://openalex.org/W2971213186","https://openalex.org/W3206893245"],"abstract_inverted_index":{"Many":[0],"recent":[1],"successful":[2],"(deep)":[3],"reinforcement":[4],"learning":[5],"algorithms":[6,99],"make":[7],"use":[8],"of":[9,23,40,62,68,79,89,93],"regularization,":[10],"generally":[11],"based":[12],"on":[13],"entropy":[14],"or":[15,111],"Kullback-Leibler":[16],"divergence.":[17],"We":[18],"propose":[19],"a":[20,37,66,76],"general":[21,46,90],"theory":[22,64],"regularized":[24,69],"Markov":[25],"Decision":[26],"Processes":[27],"that":[28],"generalizes":[29],"these":[30],"approaches":[31],"in":[32],"two":[33],"directions:":[34],"we":[35,43],"consider":[36,44],"larger":[38],"class":[39],"regularizers,":[41],"and":[42,55,72],"the":[45,73],"modified":[47],"policy":[48,53],"iteration":[49,54],"approach,":[50],"encompassing":[51],"both":[52],"value":[56],"iteration.":[57],"The":[58],"core":[59],"building":[60],"blocks":[61],"this":[63],"are":[65,115],"notion":[67],"Bellman":[70],"operator":[71],"Legendre-Fenchel":[74],"transform,":[75],"classical":[77,98],"tool":[78],"convex":[80,124],"optimization.":[81],"This":[82,118],"approach":[83],"allows":[84],"for":[85],"error":[86],"propagation":[87],"analyses":[88],"algorithmic":[91],"schemes":[92],"which":[94],"(possibly":[95],"variants":[96],"of)":[97],"such":[100],"as":[101],"Trust":[102],"Region":[103],"Policy":[104,113],"Optimization,":[105],"Soft":[106],"Q-learning,":[107],"Stochastic":[108],"Actor":[109],"Critic":[110],"Dynamic":[112],"Programming":[114],"special":[116],"cases.":[117],"also":[119],"draws":[120],"connections":[121],"to":[122,127],"proximal":[123],"optimization,":[125],"especially":[126],"Mirror":[128],"Descent.":[129]},"counts_by_year":[{"year":2025,"cited_by_count":8},{"year":2024,"cited_by_count":14},{"year":2023,"cited_by_count":14},{"year":2022,"cited_by_count":16},{"year":2021,"cited_by_count":44},{"year":2020,"cited_by_count":36},{"year":2019,"cited_by_count":13},{"year":2018,"cited_by_count":1}],"updated_date":"2026-11-04T10:54:05.112534","created_date":"2019-02-21T00:00:00"}