{"id":"https://openalex.org/W4226401482","doi":"https://doi.org/10.24963/ijcai.2022/471","title":"Understanding the Limits of Poisoning Attacks in Episodic Reinforcement Learning","display_name":"Understanding the Limits of Poisoning Attacks in Episodic Reinforcement Learning","publication_year":2022,"publication_date":"2022-07-01","ids":{"openalex":"https://openalex.org/W4226401482","doi":"https://doi.org/10.24963/ijcai.2022/471"},"language":"en","primary_location":{"id":"doi:10.24963/ijcai.2022/471","is_oa":true,"landing_page_url":"https://doi.org/10.24963/ijcai.2022/471","pdf_url":"https://www.ijcai.org/proceedings/2022/0471.pdf","source":{"id":"https://openalex.org/S4363608755","display_name":"Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"bronze","oa_url":"https://www.ijcai.org/proceedings/2022/0471.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5063692433","display_name":"Anshuka Rangi","orcid":"https://orcid.org/0000-0002-8462-7892"},"institutions":[{"id":"https://openalex.org/I36258959","display_name":"University of California San Diego","ror":"https://ror.org/0168r3w48","country_code":"US","type":"education","lineage":["https://openalex.org/I36258959"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Anshuka Rangi","raw_affiliation_strings":["University of California San Diego","University of California San Diego, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of California San Diego","institution_ids":["https://openalex.org/I36258959"]},{"raw_affiliation_string":"University of California San Diego, USA","institution_ids":["https://openalex.org/I36258959"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101434141","display_name":"Haifeng Xu","orcid":"https://orcid.org/0000-0001-6371-4906"},"institutions":[{"id":"https://openalex.org/I51556381","display_name":"University of Virginia","ror":"https://ror.org/0153tk833","country_code":"US","type":"education","lineage":["https://openalex.org/I51556381"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Haifeng Xu","raw_affiliation_strings":["University of Virginia","University of Virginia, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Virginia","institution_ids":["https://openalex.org/I51556381"]},{"raw_affiliation_string":"University of Virginia, USA","institution_ids":["https://openalex.org/I51556381"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Long Tran-Thanh","orcid":null},"institutions":[{"id":"https://openalex.org/I39555362","display_name":"University of Warwick","ror":"https://ror.org/01a77tt86","country_code":"GB","type":"education","lineage":["https://openalex.org/I39555362"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Long Tran-Thanh","raw_affiliation_strings":["University of Warwick","University of Warwick, UK"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Warwick","institution_ids":["https://openalex.org/I39555362"]},{"raw_affiliation_string":"University of Warwick, UK","institution_ids":["https://openalex.org/I39555362"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5075999338","display_name":"Massimo Franceschetti","orcid":"https://orcid.org/0000-0002-4057-8152"},"institutions":[{"id":"https://openalex.org/I2800935791","display_name":"UC San Diego Health System","ror":"https://ror.org/01kbfgm16","country_code":"US","type":"healthcare","lineage":["https://openalex.org/I2800935791"]},{"id":"https://openalex.org/I36258959","display_name":"University of California San Diego","ror":"https://ror.org/0168r3w48","country_code":"US","type":"education","lineage":["https://openalex.org/I36258959"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Massimo Franceschetti","raw_affiliation_strings":["UC San Diego","University of California San Diego, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UC San Diego","institution_ids":["https://openalex.org/I2800935791"]},{"raw_affiliation_string":"University of California San Diego, USA","institution_ids":["https://openalex.org/I36258959"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":5,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"3394","last_page":"3400"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9983999729156494,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9983999729156494,"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/T10917","display_name":"Smart Grid Security and Resilience","score":0.9945999979972839,"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/T11241","display_name":"Advanced Malware Detection Techniques","score":0.9876000285148621,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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/reinforcement-learning","display_name":"Reinforcement learning","score":0.8878284096717834},{"id":"https://openalex.org/keywords/bounded-function","display_name":"Bounded function","score":0.669876217842102},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.6563194990158081},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6527173519134521},{"id":"https://openalex.org/keywords/adversary","display_name":"Adversary","score":0.6270994544029236},{"id":"https://openalex.org/keywords/order","display_name":"Order (exchange)","score":0.5701443552970886},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.5657511949539185},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.43384161591529846},{"id":"https://openalex.org/keywords/computer-security","display_name":"Computer security","score":0.3820970952510834},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.17420312762260437},{"id":"https://openalex.org/keywords/economics","display_name":"Economics","score":0.08348625898361206}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.8878284096717834},{"id":"https://openalex.org/C34388435","wikidata":"https://www.wikidata.org/wiki/Q2267362","display_name":"Bounded function","level":2,"score":0.669876217842102},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.6563194990158081},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6527173519134521},{"id":"https://openalex.org/C41065033","wikidata":"https://www.wikidata.org/wiki/Q2825412","display_name":"Adversary","level":2,"score":0.6270994544029236},{"id":"https://openalex.org/C182306322","wikidata":"https://www.wikidata.org/wiki/Q1779371","display_name":"Order (exchange)","level":2,"score":0.5701443552970886},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5657511949539185},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.43384161591529846},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.3820970952510834},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.17420312762260437},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.08348625898361206},{"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/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0},{"id":"https://openalex.org/C10138342","wikidata":"https://www.wikidata.org/wiki/Q43015","display_name":"Finance","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}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.24963/ijcai.2022/471","is_oa":true,"landing_page_url":"https://doi.org/10.24963/ijcai.2022/471","pdf_url":"https://www.ijcai.org/proceedings/2022/0471.pdf","source":{"id":"https://openalex.org/S4363608755","display_name":"Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"},{"id":"pmh:oai:wrap.warwick.ac.uk:167181","is_oa":false,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4306400665","display_name":"Warwick Research Archive Portal (University of Warwick)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I39555362","host_organization_name":"University of Warwick","host_organization_lineage":["https://openalex.org/I39555362"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"","raw_type":"Conference Item"},{"id":"pmh:oai:arXiv.org:2208.13663","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2208.13663","pdf_url":"https://arxiv.org/pdf/2208.13663","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"}],"best_oa_location":{"id":"doi:10.24963/ijcai.2022/471","is_oa":true,"landing_page_url":"https://doi.org/10.24963/ijcai.2022/471","pdf_url":"https://www.ijcai.org/proceedings/2022/0471.pdf","source":{"id":"https://openalex.org/S4363608755","display_name":"Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"},"sustainable_development_goals":[{"score":0.5400000214576721,"display_name":"Peace, Justice and strong institutions","id":"https://metadata.un.org/sdg/16"}],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4226401482.pdf","grobid_xml":"https://content.openalex.org/works/W4226401482.grobid-xml"},"referenced_works_count":27,"referenced_works":["https://openalex.org/W2103104707","https://openalex.org/W2123990969","https://openalex.org/W2614208603","https://openalex.org/W2616841723","https://openalex.org/W2799544270","https://openalex.org/W2941205169","https://openalex.org/W2949103145","https://openalex.org/W2950808099","https://openalex.org/W2950963126","https://openalex.org/W2962755762","https://openalex.org/W2963049774","https://openalex.org/W2963068985","https://openalex.org/W2965721472","https://openalex.org/W2979429887","https://openalex.org/W3005360685","https://openalex.org/W3007968297","https://openalex.org/W3034593529","https://openalex.org/W3035388736","https://openalex.org/W3091226474","https://openalex.org/W3107230347","https://openalex.org/W3116781056","https://openalex.org/W3129608001","https://openalex.org/W3132066662","https://openalex.org/W4283800968","https://openalex.org/W4287327356","https://openalex.org/W4287328532","https://openalex.org/W4287726313"],"related_works":["https://openalex.org/W4388150944","https://openalex.org/W4242235492","https://openalex.org/W4306904969","https://openalex.org/W4237162029","https://openalex.org/W2367268135","https://openalex.org/W2138720691","https://openalex.org/W4362501864","https://openalex.org/W4380318855","https://openalex.org/W2385701518","https://openalex.org/W3084456289"],"abstract_inverted_index":{"To":[0],"understand":[1],"the":[2,30,41,50,58,90,115,175],"security":[3],"threats":[4],"to":[5,15,98,135,142,170],"reinforcement":[6],"learning":[7,19,96,140],"(RL)":[8],"algorithms,":[9],"this":[10],"paper":[11],"studies":[12],"poisoning":[13,38,165],"attacks":[14,53,129],"manipulate":[16,93,137],"any":[17,94,100,112,138,144],"order-optimal":[18,95,139],"algorithm":[20,97,141],"towards":[21],"a":[22,80],"targeted":[23,101,145],"policy":[24,102,146],"in":[25,120],"episodic":[26],"RL":[27,179],"and":[28,87,167],"examines":[29],"potential":[31],"damage":[32],"of":[33,37,43,52,114,150,177],"two":[34],"natural":[35],"types":[36],"attacks,":[39,166],"i.e.,":[40],"manipulation":[42,73,77,128],"reward":[44,66,72,86,122,127],"or":[45,62,74,160],"action.":[46],"We":[47],"discover":[48],"that":[49,70,126],"effect":[51],"crucially":[54],"depends":[55],"on":[56,174],"whether":[57],"rewards":[59],"are":[60,130,168],"bounded":[61,65],"unbounded.":[63],"In":[64,118],"settings,":[67,123],"we":[68,124],"show":[69,125],"only":[71,75],"action":[76,88],"cannot":[78,161],"guarantee":[79],"successful":[81],"attack.":[82],"However,":[83],"by":[84,164],"combining":[85],"manipulation,":[89],"adversary":[91,134],"can":[92,159],"follow":[99,143],"with":[103],"\\Theta(\\sqrt{T})":[104],"total":[105],"attack":[106],"cost,":[107],"which":[108],"is":[109],"order-optimal,":[110],"without":[111],"knowledge":[113],"underlying":[116],"MDP.":[117],"contrast,":[119],"unbounded":[121],"sufficient":[131],"for":[132],"an":[133],"successfully":[136],"using":[147],"\\tilde{O}(\\sqrt{T})":[148],"amount":[149],"contamination.":[151],"Our":[152],"results":[153],"reveal":[154],"useful":[155],"insights":[156],"about":[157],"what":[158],"be":[162],"achieved":[163],"set":[169],"spur":[171],"more":[172],"work":[173],"design":[176],"robust":[178],"algorithms.":[180]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":2},{"year":2023,"cited_by_count":2}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2022-05-05T00:00:00"}