{"id":"https://openalex.org/W2602963933","doi":"https://doi.org/10.48550/arxiv.1703.02702","title":"Robust Adversarial Reinforcement Learning","display_name":"Robust Adversarial Reinforcement Learning","publication_year":2017,"publication_date":"2017-03-08","ids":{"openalex":"https://openalex.org/W2602963933","doi":"https://doi.org/10.48550/arxiv.1703.02702","mag":"2602963933"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:1703.02702","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/1703.02702","pdf_url":"https://arxiv.org/pdf/1703.02702","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/1703.02702","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5103895643","display_name":"Lerrel Pinto","orcid":"https://orcid.org/0000-0002-2726-0633","observed_orcids":["https://orcid.org/0000-0002-2726-0633"]},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Pinto, Lerrel","raw_affiliation_strings":["Carnegie Mellon University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5021713982","display_name":"James Davidson","orcid":"https://orcid.org/0000-0001-5853-4310","observed_orcids":["https://orcid.org/0000-0001-5853-4310"]},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Davidson, James","raw_affiliation_strings":["[Google Brain]"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"[Google Brain]","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066107323","display_name":"Rahul Sukthankar","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Sukthankar, Rahul","raw_affiliation_strings":["Google Research,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google Research,","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101761266","display_name":"Abhinav Gupta","orcid":"https://orcid.org/0000-0002-9197-0736","observed_orcids":["https://orcid.org/0000-0002-9197-0736","https://orcid.org/0000-0002-3646-2421"]},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]},{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Gupta, Abhinav","raw_affiliation_strings":["Carnegie Mellon University and Google Research#TAB#"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University and Google Research#TAB#","institution_ids":["https://openalex.org/I1291425158","https://openalex.org/I74973139"]}]}],"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":557,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"2817","last_page":"2826"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":1.0,"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":1.0,"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":9.999999747378752e-05,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.0,"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/h-infinity-control","display_name":"H-infinity control","score":0.765999972820282},{"id":"https://openalex.org/keywords/robustness-to-modeling-errors","display_name":"robustness to modeling errors","score":0.6110000014305115},{"id":"https://openalex.org/keywords/robustness-to-disturbances","display_name":"robustness to disturbances","score":0.42100000381469727},{"id":"https://openalex.org/keywords/zero-sum-games","display_name":"zero-sum games","score":0.5260000228881836},{"id":"https://openalex.org/keywords/training-stability","display_name":"training stability","score":0.4569999873638153},{"id":"https://openalex.org/keywords/deep-reinforcement-learning","display_name":"deep reinforcement learning","score":0.4440000057220459}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.8467103242874146},{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.7961954474449158},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7186185121536255},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6471161842346191},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.6346005797386169},{"id":"https://openalex.org/C41065033","wikidata":"https://www.wikidata.org/wiki/Q2825412","display_name":"Adversary","level":2,"score":0.586288332939148},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.5722438097000122},{"id":"https://openalex.org/C149728462","wikidata":"https://www.wikidata.org/wiki/Q751319","display_name":"Minimax","level":2,"score":0.5437232255935669},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4709887206554413},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4549800753593445},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.30264604091644287},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.12586024403572083},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.11689135432243347},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0}],"mesh":[],"locations_count":3,"locations":[{"id":"pmh:oai:arXiv.org:1703.02702","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/1703.02702","pdf_url":"https://arxiv.org/pdf/1703.02702","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":"doi:10.48550/arxiv.1703.02702","doi":"https://doi.org/10.48550/arxiv.1703.02702","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.1703.02702","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"},{"id":"mag:2602963933","doi":null,"is_oa":false,"landing_page_url":"http://proceedings.mlr.press/v70/pinto17a/pinto17a.pdf","pdf_url":null,"source":{"id":"https://openalex.org/S4306419644","display_name":"International Conference on Machine Learning","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":"International Conference on Machine Learning","raw_type":null}],"best_oa_location":{"id":"pmh:oai:arXiv.org:1703.02702","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/1703.02702","pdf_url":"https://arxiv.org/pdf/1703.02702","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":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W2602963933.pdf","grobid_xml":"https://content.openalex.org/works/W2602963933.grobid-xml"},"referenced_works_count":35,"referenced_works":["https://openalex.org/W206679605","https://openalex.org/W1542941925","https://openalex.org/W1550386171","https://openalex.org/W1554705102","https://openalex.org/W1579118835","https://openalex.org/W1771410628","https://openalex.org/W1788877992","https://openalex.org/W2041357003","https://openalex.org/W2107726111","https://openalex.org/W2119717200","https://openalex.org/W2130801532","https://openalex.org/W2145339207","https://openalex.org/W2158782408","https://openalex.org/W2165622730","https://openalex.org/W2167696704","https://openalex.org/W2173248099","https://openalex.org/W2186601217","https://openalex.org/W2257979135","https://openalex.org/W2290354866","https://openalex.org/W2291973609","https://openalex.org/W2291986326","https://openalex.org/W2294316001","https://openalex.org/W2341497066","https://openalex.org/W2411541852","https://openalex.org/W2528973876","https://openalex.org/W2529477964","https://openalex.org/W2530944449","https://openalex.org/W2534269850","https://openalex.org/W2553303224","https://openalex.org/W2580175322","https://openalex.org/W2963207607","https://openalex.org/W2963614114","https://openalex.org/W2965916140","https://openalex.org/W3037207827","https://openalex.org/W6998967213"],"related_works":["https://openalex.org/W7135226595","https://openalex.org/W3158314496","https://openalex.org/W4309326691","https://openalex.org/W2964108292","https://openalex.org/W4403094434","https://openalex.org/W7214211367","https://openalex.org/W4283797804","https://openalex.org/W4415015574","https://openalex.org/W2773691349","https://openalex.org/W2967292964"],"abstract_inverted_index":{"Deep":[0],"neural":[1],"networks":[2],"coupled":[3],"with":[4],"fast":[5],"simulation":[6,35],"and":[7,36,89,93,172,191],"improved":[8],"computation":[9],"have":[10],"led":[11],"to":[12,28,46,62,67,72,121,134,186],"recent":[13],"successes":[14],"in":[15,55,91,102,123,165,188,197],"the":[16,32,58,103,108,124,135,154,194,198,201],"field":[17],"of":[18,110,126,200],"reinforcement":[19,113],"learning":[20,52,114,156],"(RL).":[21],"However,":[22],"most":[23],"current":[24],"RL-based":[25],"approaches":[26,44],"fail":[27,45],"generalize":[29],"since:":[30],"(a)":[31,179],"gap":[33],"between":[34],"real":[37,56],"world":[38],"is":[39,53,141,184],"so":[40],"large":[41],"that":[42,85,130,144,176],"policy-learning":[43],"transfer;":[47],"(b)":[48,183],"even":[49,196],"if":[50],"policy":[51,155],"done":[54],"world,":[57],"data":[59],"scarcity":[60],"leads":[61],"failed":[63],"generalization":[64],"from":[65,79],"training":[66,92,181],"test":[68,94],"scenarios":[69,95],"(e.g.,":[70],"due":[71],"different":[73],"friction":[74],"or":[75],"object":[76],"masses).":[77],"Inspired":[78],"H-infinity":[80],"control":[81],"methods,":[82],"we":[83,117],"note":[84],"both":[86],"modeling":[87],"errors":[88],"differences":[90,187],"can":[96],"be":[97],"viewed":[98],"as":[99,157],"extra":[100],"forces/disturbances":[101],"system.":[104,136],"This":[105],"paper":[106],"proposes":[107],"idea":[109],"robust":[111,185],"adversarial":[112],"(RARL),":[115],"where":[116],"train":[118],"an":[119,148],"agent":[120],"operate":[122],"presence":[125],"a":[127,158],"destabilizing":[128],"adversary":[129,140],"applies":[131],"disturbance":[132],"forces":[133],"The":[137],"jointly":[138],"trained":[139],"reinforced":[142],"--":[143],"is,":[145],"it":[146],"learns":[147],"optimal":[149],"destabilization":[150],"policy.":[151],"We":[152],"formulate":[153],"zero-sum,":[159],"minimax":[160],"objective":[161],"function.":[162],"Extensive":[163],"experiments":[164],"multiple":[166],"environments":[167],"(InvertedPendulum,":[168],"HalfCheetah,":[169],"Swimmer,":[170],"Hopper":[171],"Walker2d)":[173],"conclusively":[174],"demonstrate":[175],"our":[177],"method":[178],"improves":[180],"stability;":[182],"training/test":[189],"conditions;":[190],"c)":[192],"outperform":[193],"baseline":[195],"absence":[199],"adversary.":[202]},"counts_by_year":[{"year":2026,"cited_by_count":8},{"year":2025,"cited_by_count":30},{"year":2024,"cited_by_count":77},{"year":2023,"cited_by_count":75},{"year":2022,"cited_by_count":74},{"year":2021,"cited_by_count":115},{"year":2020,"cited_by_count":84},{"year":2019,"cited_by_count":65},{"year":2018,"cited_by_count":24},{"year":2017,"cited_by_count":5}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2017-04-07T00:00:00"}