{"id":"https://openalex.org/W4414359274","doi":"https://doi.org/10.24963/ijcai.2025/588","title":"PNAct: Crafting Backdoor Attacks in Safe Reinforcement Learning","display_name":"PNAct: Crafting Backdoor Attacks in Safe Reinforcement Learning","publication_year":2025,"publication_date":"2025-09-01","ids":{"openalex":"https://openalex.org/W4414359274","doi":"https://doi.org/10.24963/ijcai.2025/588"},"language":"en","primary_location":{"id":"doi:10.24963/ijcai.2025/588","is_oa":false,"landing_page_url":"https://doi.org/10.24963/ijcai.2025/588","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"},"type":"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/A5101303108","display_name":"Weiran Guo","orcid":"https://orcid.org/0009-0008-1626-8745"},"institutions":[{"id":"https://openalex.org/I116953780","display_name":"Tongji University","ror":"https://ror.org/03rc6as71","country_code":"CN","type":"education","lineage":["https://openalex.org/I116953780"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Weiran Guo","raw_affiliation_strings":["Tongji University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tongji University","institution_ids":["https://openalex.org/I116953780"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5050572529","display_name":"Guanjun Liu","orcid":"https://orcid.org/0000-0002-3654-6908"},"institutions":[{"id":"https://openalex.org/I116953780","display_name":"Tongji University","ror":"https://ror.org/03rc6as71","country_code":"CN","type":"education","lineage":["https://openalex.org/I116953780"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Guanjun Liu","raw_affiliation_strings":["Tongji University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tongji University","institution_ids":["https://openalex.org/I116953780"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5055818068","display_name":"Ziyuan Zhou","orcid":"https://orcid.org/0000-0002-0835-2751"},"institutions":[{"id":"https://openalex.org/I116953780","display_name":"Tongji University","ror":"https://ror.org/03rc6as71","country_code":"CN","type":"education","lineage":["https://openalex.org/I116953780"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ziyuan Zhou","raw_affiliation_strings":["Tongji University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tongji University","institution_ids":["https://openalex.org/I116953780"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100398654","display_name":"Ling Wang","orcid":"https://orcid.org/0000-0001-7901-3995"},"institutions":[{"id":"https://openalex.org/I116953780","display_name":"Tongji University","ror":"https://ror.org/03rc6as71","country_code":"CN","type":"education","lineage":["https://openalex.org/I116953780"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ling Wang","raw_affiliation_strings":["Tongji University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tongji University","institution_ids":["https://openalex.org/I116953780"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":3.5175,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.93422255,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":91,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"5280","last_page":"5288"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.8669999837875366,"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":0.8669999837875366,"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/T11241","display_name":"Advanced Malware Detection Techniques","score":0.8282999992370605,"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/backdoor","display_name":"Backdoor","score":0.98089998960495},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.6524999737739563},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.5922999978065491},{"id":"https://openalex.org/keywords/metric","display_name":"Metric (unit)","score":0.4618000090122223},{"id":"https://openalex.org/keywords/safer","display_name":"SAFER","score":0.4050000011920929},{"id":"https://openalex.org/keywords/field","display_name":"Field (mathematics)","score":0.3950999975204468},{"id":"https://openalex.org/keywords/sample","display_name":"Sample (material)","score":0.38499999046325684}],"concepts":[{"id":"https://openalex.org/C2781045450","wikidata":"https://www.wikidata.org/wiki/Q254569","display_name":"Backdoor","level":2,"score":0.98089998960495},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6524999737739563},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6288999915122986},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.6262999773025513},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.5922999978065491},{"id":"https://openalex.org/C176217482","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Metric (unit)","level":2,"score":0.4618000090122223},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.4332999885082245},{"id":"https://openalex.org/C2776654903","wikidata":"https://www.wikidata.org/wiki/Q2601463","display_name":"SAFER","level":2,"score":0.4050000011920929},{"id":"https://openalex.org/C9652623","wikidata":"https://www.wikidata.org/wiki/Q190109","display_name":"Field (mathematics)","level":2,"score":0.3950999975204468},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.38499999046325684},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.383899986743927},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.337799996137619},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3296999931335449},{"id":"https://openalex.org/C67203356","wikidata":"https://www.wikidata.org/wiki/Q1321905","display_name":"Reinforcement","level":2,"score":0.32260000705718994},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.27079999446868896},{"id":"https://openalex.org/C2777610993","wikidata":"https://www.wikidata.org/wiki/Q3433064","display_name":"Riprap","level":2,"score":0.2700999975204468},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.259799987077713},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.251800000667572}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.24963/ijcai.2025/588","is_oa":false,"landing_page_url":"https://doi.org/10.24963/ijcai.2025/588","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Reinforcement":[0,22],"Learning":[1,23],"(RL)":[2],"is":[3,51,80,99],"widely":[4],"used":[5],"in":[6,76,85],"tasks":[7],"where":[8,103],"agents":[9,36,59],"interact":[10],"with":[11,148,159],"an":[12,129],"environment":[13],"to":[14,38,53,100,116,136],"maximize":[15],"rewards.":[16],"Building":[17],"on":[18],"this":[19,44],"foundation,":[20],"Safe":[21,49,77,87,160],"(Safe":[24],"RL)":[25],"incorporates":[26],"a":[27],"cost":[28],"metric":[29],"alongside":[30],"the":[31,67,81,86,123,138,149,155,164],"reward":[32],"metric,":[33],"ensuring":[34],"that":[35,48,90],"adhere":[37],"safety":[39],"constraints":[40],"during":[41],"decision-making.":[42],"In":[43],"paper,":[45],"we":[46,65,133],"identify":[47],"RL":[50,88,161],"vulnerable":[52],"backdoor":[54,74,143],"attacks,":[55],"which":[56],"can":[57],"manipulate":[58],"into":[60],"performing":[61],"unsafe":[62],"actions.":[63],"First,":[64],"introduce":[66],"relevant":[68],"concepts":[69],"and":[70,94,110,127,162,171],"evaluation":[71],"metrics":[72],"for":[73],"attacks":[75],"RL.":[78],"It":[79],"first":[82],"attack":[83,130,144],"framework":[84],"field":[89],"involves":[91],"both":[92],"Positive":[93],"Negative":[95],"Action":[96],"sample":[97],"(PNAct)":[98],"implant":[101],"backdoors,":[102],"positive":[104],"action":[105,112],"samples":[106,113],"provide":[107],"reference":[108],"actions":[109,115],"negative":[111],"indicate":[114],"be":[117],"avoided.":[118],"We":[119],"theoretically":[120],"point":[121],"out":[122],"properties":[124],"of":[125,140,166],"PNAct":[126],"design":[128],"algorithm.":[131],"Finally,":[132],"conduct":[134],"experiments":[135],"evaluate":[137],"effectiveness":[139],"our":[141],"proposed":[142],"framework,":[145],"evaluating":[146],"it":[147],"established":[150],"metrics.":[151],"This":[152],"paper":[153],"highlights":[154],"potential":[156],"risks":[157],"associated":[158],"underscores":[163],"feasibility":[165],"such":[167],"attacks.":[168],"Our":[169],"code":[170],"supplementary":[172],"material":[173],"are":[174],"available":[175],"at":[176],"https://github.com/azure-123/PNAct.":[177]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}