{"id":"https://openalex.org/W2959399500","doi":"https://doi.org/10.1177/0278364920959444","title":"PAC-Bayes control: learning policies that provably generalize to novel environments","display_name":"PAC-Bayes control: learning policies that provably generalize to novel environments","publication_year":2020,"publication_date":"2020-10-03","ids":{"openalex":"https://openalex.org/W2959399500","doi":"https://doi.org/10.1177/0278364920959444","mag":"2959399500"},"language":"en","primary_location":{"id":"doi:10.1177/0278364920959444","is_oa":false,"landing_page_url":"https://doi.org/10.1177/0278364920959444","pdf_url":null,"source":{"id":"https://openalex.org/S73484101","display_name":"The International Journal of Robotics Research","issn_l":"0278-3649","issn":["0278-3649","1741-3176"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320017","host_organization_name":"SAGE Publishing","host_organization_lineage":["https://openalex.org/P4310320017"],"host_organization_lineage_names":["SAGE Publishing"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"The International Journal of Robotics Research","raw_type":"journal-article"},"type":"preprint","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/1806.04225","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5102792178","display_name":"Anirudha Majumdar","orcid":"https://orcid.org/0009-0002-2296-7485"},"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":"Anirudha Majumdar","raw_affiliation_strings":["Department of Mechanical and Aerospace Engineering,Princeton University, Princeton, NJ, USA","Department of Mechanical and Aerospace Engineering, Princeton University , Princeton, NJ, USA"],"affiliations":[{"raw_affiliation_string":"Department of Mechanical and Aerospace Engineering,Princeton University, Princeton, NJ, USA","institution_ids":["https://openalex.org/I20089843"]},{"raw_affiliation_string":"Department of Mechanical and Aerospace Engineering, Princeton University , Princeton, NJ, USA","institution_ids":["https://openalex.org/I20089843"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5110782445","display_name":"Alec Farid","orcid":"https://orcid.org/0000-0003-4253-496X"},"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":"Alec Farid","raw_affiliation_strings":["Department of Mechanical and Aerospace Engineering,Princeton University, Princeton, NJ, USA","Department of Mechanical and Aerospace Engineering, Princeton University , Princeton, NJ, USA"],"affiliations":[{"raw_affiliation_string":"Department of Mechanical and Aerospace Engineering,Princeton University, Princeton, NJ, USA","institution_ids":["https://openalex.org/I20089843"]},{"raw_affiliation_string":"Department of Mechanical and Aerospace Engineering, Princeton University , Princeton, NJ, USA","institution_ids":["https://openalex.org/I20089843"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5069556905","display_name":"Anoopkumar Sonar","orcid":"https://orcid.org/0000-0003-0478-0254"},"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":"Anoopkumar Sonar","raw_affiliation_strings":["Department of Computer Science, Princeton University,Princeton, NJ, USA","Princeton University"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science, Princeton University,Princeton, NJ, USA","institution_ids":["https://openalex.org/I20089843"]},{"raw_affiliation_string":"Princeton University","institution_ids":["https://openalex.org/I20089843"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5102792178"],"corresponding_institution_ids":["https://openalex.org/I20089843"],"apc_list":null,"apc_paid":null,"fwci":0.5485,"has_fulltext":true,"cited_by_count":4,"citation_normalized_percentile":{"value":0.72169054,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":96},"biblio":{"volume":"40","issue":"2-3","first_page":"574","last_page":"593"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.9988999962806702,"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":0.9988999962806702,"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":0.9976999759674072,"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/T12072","display_name":"Machine Learning and Algorithms","score":0.9973999857902527,"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":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6957352757453918},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.5891954898834229},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.5860812664031982},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5704004764556885},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.5627954006195068},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.48612427711486816},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.4825815260410309},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.45592737197875977},{"id":"https://openalex.org/keywords/stochastic-control","display_name":"Stochastic control","score":0.4503341317176819},{"id":"https://openalex.org/keywords/optimal-control","display_name":"Optimal control","score":0.3722023367881775},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.18192440271377563}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6957352757453918},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.5891954898834229},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.5860812664031982},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5704004764556885},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.5627954006195068},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.48612427711486816},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4825815260410309},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.45592737197875977},{"id":"https://openalex.org/C170131372","wikidata":"https://www.wikidata.org/wiki/Q7617811","display_name":"Stochastic control","level":3,"score":0.4503341317176819},{"id":"https://openalex.org/C91575142","wikidata":"https://www.wikidata.org/wiki/Q1971426","display_name":"Optimal control","level":2,"score":0.3722023367881775},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.18192440271377563},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.1177/0278364920959444","is_oa":false,"landing_page_url":"https://doi.org/10.1177/0278364920959444","pdf_url":null,"source":{"id":"https://openalex.org/S73484101","display_name":"The International Journal of Robotics Research","issn_l":"0278-3649","issn":["0278-3649","1741-3176"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320017","host_organization_name":"SAGE Publishing","host_organization_lineage":["https://openalex.org/P4310320017"],"host_organization_lineage_names":["SAGE Publishing"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"The International Journal of Robotics Research","raw_type":"journal-article"},{"id":"pmh:oai:arXiv.org:1806.04225","is_oa":true,"landing_page_url":"http://arxiv.org/abs/1806.04225","pdf_url":"https://arxiv.org/pdf/1806.04225","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":"mag:2959399500","is_oa":true,"landing_page_url":"http://export.arxiv.org/pdf/1806.04225","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":"arXiv (Cornell University)","raw_type":null},{"id":"doi:10.48550/arxiv.1806.04225","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.1806.04225","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:1806.04225","is_oa":true,"landing_page_url":"http://arxiv.org/abs/1806.04225","pdf_url":"https://arxiv.org/pdf/1806.04225","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":[{"id":"https://metadata.un.org/sdg/17","score":0.44999998807907104,"display_name":"Partnerships for the goals"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320309480","display_name":"Nvidia","ror":"https://ror.org/03jdj4y14"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W2959399500.pdf","grobid_xml":"https://content.openalex.org/works/W2959399500.grobid-xml"},"referenced_works_count":81,"referenced_works":["https://openalex.org/W60181528","https://openalex.org/W1144593952","https://openalex.org/W1496462336","https://openalex.org/W1500134124","https://openalex.org/W1505937442","https://openalex.org/W1622633983","https://openalex.org/W1628114069","https://openalex.org/W1632601927","https://openalex.org/W1771410628","https://openalex.org/W1792768639","https://openalex.org/W1811750039","https://openalex.org/W1846446278","https://openalex.org/W1976144763","https://openalex.org/W1980969546","https://openalex.org/W1989125340","https://openalex.org/W1999156278","https://openalex.org/W2014384147","https://openalex.org/W2049606892","https://openalex.org/W2095705004","https://openalex.org/W2099089474","https://openalex.org/W2108598243","https://openalex.org/W2114396319","https://openalex.org/W2115065944","https://openalex.org/W2116314036","https://openalex.org/W2129516068","https://openalex.org/W2130105540","https://openalex.org/W2131024815","https://openalex.org/W2132004142","https://openalex.org/W2136144249","https://openalex.org/W2138263042","https://openalex.org/W2139338362","https://openalex.org/W2141698884","https://openalex.org/W2149991487","https://openalex.org/W2150434953","https://openalex.org/W2159555929","https://openalex.org/W2161521419","https://openalex.org/W2165265965","https://openalex.org/W2166944917","https://openalex.org/W2168022998","https://openalex.org/W2168359464","https://openalex.org/W2170207925","https://openalex.org/W2190691619","https://openalex.org/W2221045814","https://openalex.org/W2288174618","https://openalex.org/W2290452516","https://openalex.org/W2296319761","https://openalex.org/W2297172695","https://openalex.org/W2321429733","https://openalex.org/W2489939061","https://openalex.org/W2593841437","https://openalex.org/W2604117713","https://openalex.org/W2604763608","https://openalex.org/W2619268125","https://openalex.org/W2734960757","https://openalex.org/W2737856893","https://openalex.org/W2766108544","https://openalex.org/W2777005400","https://openalex.org/W2777629900","https://openalex.org/W2788800397","https://openalex.org/W2808177011","https://openalex.org/W2886067286","https://openalex.org/W2949117887","https://openalex.org/W2949608212","https://openalex.org/W2950220847","https://openalex.org/W2951660448","https://openalex.org/W2953249127","https://openalex.org/W2962822210","https://openalex.org/W2962835968","https://openalex.org/W2962887844","https://openalex.org/W2963285844","https://openalex.org/W2963695615","https://openalex.org/W2963739978","https://openalex.org/W2963956866","https://openalex.org/W2964068654","https://openalex.org/W2964112890","https://openalex.org/W2964161785","https://openalex.org/W2964248288","https://openalex.org/W4243316134","https://openalex.org/W4243551976","https://openalex.org/W4244569433","https://openalex.org/W4250589301"],"related_works":["https://openalex.org/W3011338904","https://openalex.org/W3006648316","https://openalex.org/W2134153324","https://openalex.org/W3011651653","https://openalex.org/W2789561913","https://openalex.org/W3100191104","https://openalex.org/W2182877511","https://openalex.org/W2912212713","https://openalex.org/W46329919","https://openalex.org/W3034514558","https://openalex.org/W2169498096","https://openalex.org/W2963937357","https://openalex.org/W3128863975","https://openalex.org/W2098284983","https://openalex.org/W3087653545","https://openalex.org/W2592309595","https://openalex.org/W3137841177","https://openalex.org/W3133308194","https://openalex.org/W2964136501","https://openalex.org/W2805099615"],"abstract_inverted_index":{"Our":[0,201],"goal":[1],"is":[2,29],"to":[3,13,30,58,83,111,173,209],"learn":[4],"control":[5,56,98],"policies":[6,57,99,151,179],"for":[7,191,214],"robots":[8],"that":[9,87,108],"provably":[10],"generalize":[11],"well":[12],"novel":[14,59,101],"environments":[15,60],"given":[16],"a":[17,41,51,139],"dataset":[18],"of":[19,50,55,63,96,148,169,206],"example":[20],"environments.":[21,102,200],"The":[22,116],"key":[23],"technical":[24],"idea":[25],"behind":[26],"our":[27,170,207],"approach":[28,171,208],"leverage":[31],"tools":[32],"from":[33],"generalization":[34,54,62,212],"theory":[35],"in":[36,47,65,129,131],"machine":[37],"learning":[38,68,106,174],"by":[39],"exploiting":[40],"precise":[42],"analogy":[43],"(which":[44],"we":[45,72,135,156],"present":[46,166,188],"the":[48,66,74,93,132,144,192,204],"form":[49],"reduction)":[52],"between":[53],"and":[61,180,220,233],"hypotheses":[64],"supervised":[67],"setting.":[69],"In":[70,143],"particular,":[71],"utilize":[73],"probably":[75],"approximately":[76],"correct":[77],"(PAC)-Bayes":[78],"framework,":[79],"which":[80],"allows":[81],"us":[82],"obtain":[84],"upper":[85,114,159],"bounds":[86],"hold":[88],"with":[89,217],"high":[90],"probability":[91],"on":[92],"expected":[94],"cost":[95],"(stochastic)":[97],"across":[100],"We":[103,165,186],"propose":[104],"policy":[105,141],"algorithms":[107],"explicitly":[109],"seek":[110],"minimize":[112,157],"this":[113,158],"bound.":[115],"corresponding":[117],"optimization":[118,125],"problem":[119],"can":[120],"be":[121],"solved":[122],"using":[123,161],"convex":[124],"(relative":[126],"entropy":[127],"programming":[128],"particular)":[130],"setting":[133,147],"where":[134],"are":[136],"optimizing":[137],"over":[138],"finite":[140],"space.":[142],"more":[145],"general":[146],"continuously":[149],"parameterized":[150],"(e.g.,":[152,224,230],"neural":[153,182,234],"network":[154],"policies),":[155],"bound":[160],"stochastic":[162],"gradient":[163],"descent.":[164],"simulated":[167],"results":[168,190],"applied":[172],"(1)":[175],"reactive":[176],"obstacle":[177,199],"avoidance":[178],"(2)":[181],"network-based":[183,235],"grasping":[184],"policies.":[185,236],"also":[187],"hardware":[189],"Parrot":[193],"Swing":[194],"drone":[195],"navigating":[196],"through":[197],"different":[198],"examples":[202],"demonstrate":[203],"potential":[205],"provide":[210],"strong":[211],"guarantees":[213],"robotic":[215],"systems":[216],"continuous":[218],"state":[219],"action":[221],"spaces,":[222],"complicated":[223],"nonlinear)":[225],"dynamics,":[226],"rich":[227],"sensory":[228],"inputs":[229],"depth":[231],"images),":[232]},"counts_by_year":[{"year":2022,"cited_by_count":1},{"year":2021,"cited_by_count":1},{"year":2020,"cited_by_count":2}],"updated_date":"2026-03-20T23:20:44.827607","created_date":"2025-10-10T00:00:00"}