{"id":"https://openalex.org/W4416749731","doi":"https://doi.org/10.1109/iros60139.2025.11246652","title":"Robust Deep Reinforcement Learning in Robotics via Adaptive Gradient-Masked Adversarial Attacks","display_name":"Robust Deep Reinforcement Learning in Robotics via Adaptive Gradient-Masked Adversarial Attacks","publication_year":2025,"publication_date":"2025-10-19","ids":{"openalex":"https://openalex.org/W4416749731","doi":"https://doi.org/10.1109/iros60139.2025.11246652"},"language":null,"primary_location":{"id":"doi:10.1109/iros60139.2025.11246652","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iros60139.2025.11246652","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)","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/A5080465399","display_name":"Zongyuan Zhang","orcid":"https://orcid.org/0000-0003-4577-2613"},"institutions":[{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":true,"raw_author_name":"Zongyuan Zhang","raw_affiliation_strings":["The University of Hong Kong,Department of Computer Science,Hong Kong,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The University of Hong Kong,Department of Computer Science,Hong Kong,China","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5052900159","display_name":"Tianyang Duan","orcid":null},"institutions":[{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Tianyang Duan","raw_affiliation_strings":["The University of Hong Kong,Department of Computer Science,Hong Kong,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The University of Hong Kong,Department of Computer Science,Hong Kong,China","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5089094818","display_name":"Zheng Lin","orcid":"https://orcid.org/0000-0002-4463-5652"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zheng Lin","raw_affiliation_strings":["Fudan University,School of Computer Science,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University,School of Computer Science,Shanghai,China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100308056","display_name":"Dong Huang","orcid":"https://orcid.org/0000-0001-9715-6922"},"institutions":[{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Dong Huang","raw_affiliation_strings":["The University of Hong Kong,Department of Computer Science,Hong Kong,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The University of Hong Kong,Department of Computer Science,Hong Kong,China","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5057230946","display_name":"Zihan Fang","orcid":"https://orcid.org/0000-0002-0844-9879"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zihan Fang","raw_affiliation_strings":["Fudan University,School of Computer Science,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University,School of Computer Science,Shanghai,China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5019447246","display_name":"Z. T. Sun","orcid":"https://orcid.org/0000-0003-0269-7940"},"institutions":[{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Zekai Sun","raw_affiliation_strings":["The University of Hong Kong,Department of Computer Science,Hong Kong,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The University of Hong Kong,Department of Computer Science,Hong Kong,China","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5110378574","display_name":"Ling Xiong","orcid":null},"institutions":[{"id":"https://openalex.org/I102345215","display_name":"Xihua University","ror":"https://ror.org/04gwtvf26","country_code":"CN","type":"education","lineage":["https://openalex.org/I102345215"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ling Xiong","raw_affiliation_strings":["Xihua University,School of Computer and Software Engineering,Chengdu,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Xihua University,School of Computer and Software Engineering,Chengdu,China","institution_ids":["https://openalex.org/I102345215"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100784667","display_name":"H. Liang","orcid":"https://orcid.org/0000-0002-3972-2049"},"institutions":[{"id":"https://openalex.org/I4800084","display_name":"Southwest Jiaotong University","ror":"https://ror.org/00hn7w693","country_code":"CN","type":"education","lineage":["https://openalex.org/I4800084"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hongbin Liang","raw_affiliation_strings":["Southwest Jiaotong University,School of Transportation and Logistics,Chengdu,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Southwest Jiaotong University,School of Transportation and Logistics,Chengdu,China","institution_ids":["https://openalex.org/I4800084"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015360935","display_name":"Heming Cui","orcid":"https://orcid.org/0000-0001-7746-440X"},"institutions":[{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Heming Cui","raw_affiliation_strings":["The University of Hong Kong,Department of Computer Science,Hong Kong,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The University of Hong Kong,Department of Computer Science,Hong Kong,China","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5091090025","display_name":"Yong Cui","orcid":"https://orcid.org/0000-0002-5171-739X"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yong Cui","raw_affiliation_strings":["Tsinghua University,Department of Computer Science and Technology,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tsinghua University,Department of Computer Science and Technology,Beijing,China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100602494","display_name":"Yue Gao","orcid":"https://orcid.org/0000-0002-4971-590X"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yue Gao","raw_affiliation_strings":["Fudan University,School of Computer Science,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University,School of Computer Science,Shanghai,China","institution_ids":["https://openalex.org/I24943067"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":11,"corresponding_author_ids":["https://openalex.org/A5080465399"],"corresponding_institution_ids":["https://openalex.org/I889458895"],"apc_list":null,"apc_paid":null,"fwci":6.3783,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.96535282,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"20418","last_page":"20425"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9908000230789185,"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":0.9908000230789185,"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":0.0013000000035390258,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.0010000000474974513,"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/adversarial-system","display_name":"Adversarial system","score":0.8672000169754028},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.7408999800682068},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.6229000091552734},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.524399995803833},{"id":"https://openalex.org/keywords/limiting","display_name":"Limiting","score":0.4226999878883362},{"id":"https://openalex.org/keywords/robotics","display_name":"Robotics","score":0.42239999771118164},{"id":"https://openalex.org/keywords/vulnerability","display_name":"Vulnerability (computing)","score":0.39910000562667847}],"concepts":[{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.8672000169754028},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.7408999800682068},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6833999752998352},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6807000041007996},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.6229000091552734},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.524399995803833},{"id":"https://openalex.org/C188198153","wikidata":"https://www.wikidata.org/wiki/Q1613840","display_name":"Limiting","level":2,"score":0.4226999878883362},{"id":"https://openalex.org/C34413123","wikidata":"https://www.wikidata.org/wiki/Q170978","display_name":"Robotics","level":3,"score":0.42239999771118164},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.41449999809265137},{"id":"https://openalex.org/C95713431","wikidata":"https://www.wikidata.org/wiki/Q631425","display_name":"Vulnerability (computing)","level":2,"score":0.39910000562667847},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.37619999051094055},{"id":"https://openalex.org/C48103436","wikidata":"https://www.wikidata.org/wiki/Q599031","display_name":"State (computer science)","level":2,"score":0.35260000824928284},{"id":"https://openalex.org/C89611455","wikidata":"https://www.wikidata.org/wiki/Q6804646","display_name":"Mechanism (biology)","level":2,"score":0.35179999470710754},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.3427000045776367},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.33320000767707825},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.2526000142097473}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/iros60139.2025.11246652","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iros60139.2025.11246652","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)","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":18,"referenced_works":["https://openalex.org/W1904826605","https://openalex.org/W2774644650","https://openalex.org/W2941205169","https://openalex.org/W2951360122","https://openalex.org/W2962847335","https://openalex.org/W2969542116","https://openalex.org/W3127807678","https://openalex.org/W3164563754","https://openalex.org/W3200885897","https://openalex.org/W3202281277","https://openalex.org/W3211999566","https://openalex.org/W4313126090","https://openalex.org/W4362672763","https://openalex.org/W4390871715","https://openalex.org/W4391248691","https://openalex.org/W4403422358","https://openalex.org/W4404572693","https://openalex.org/W4408354938"],"related_works":[],"abstract_inverted_index":{"Deep":[0],"reinforcement":[1],"learning":[2],"(DRL)":[3],"has":[4],"emerged":[5],"as":[6,41],"a":[7,70,78,105],"promising":[8],"approach":[9],"for":[10],"robotic":[11],"control,":[12],"but":[13],"its":[14,21],"real-world":[15],"deployment":[16],"remains":[17],"challenging":[18],"due":[19],"to":[20,23,36,83,97,109],"vulnerability":[22],"environmental":[24],"perturbations.":[25],"Existing":[26],"white-box":[27,71],"adversarial":[28,91,123,141],"attack":[29,72,124],"methods,":[30],"adapted":[31],"from":[32],"supervised":[33],"learning,":[34],"fail":[35],"effectively":[37],"target":[38],"DRL":[39,76],"agents":[40],"they":[42],"overlook":[43],"temporal":[44],"dynamics":[45],"and":[46,89,103,112,134],"indiscriminately":[47],"perturb":[48],"all":[49],"state":[50,87,101],"dimensions,":[51],"limiting":[52],"their":[53],"impact":[54],"on":[55],"long-term":[56],"rewards.":[57],"To":[58],"address":[59],"these":[60],"challenges,":[61],"we":[62],"propose":[63],"the":[64,98,128,131,136],"Adaptive":[65],"Gradient-Masked":[66],"Reinforcement":[67],"(AGMR)":[68],"Attack,":[69],"method":[73],"that":[74,119],"combines":[75],"with":[77],"gradient-based":[79],"soft":[80],"masking":[81],"mechanism":[82,108],"dynamically":[84],"identify":[85],"critical":[86],"dimensions":[88],"optimize":[90],"policies.":[92],"AGMR":[93,120],"selectively":[94],"allocates":[95],"perturbations":[96],"most":[99],"impactful":[100],"features":[102],"incorporates":[104],"dynamic":[106],"adjustment":[107],"balance":[110],"exploration":[111],"exploitation":[113],"during":[114],"training.":[115],"Extensive":[116],"experiments":[117],"demonstrate":[118],"outperforms":[121],"state-of-the-art":[122],"methods":[125],"in":[126],"degrading":[127],"performance":[129],"of":[130],"victim":[132,137],"agent":[133],"enhances":[135],"agent\u2019s":[138],"robustness":[139],"through":[140],"defense":[142],"mechanisms.":[143]},"counts_by_year":[{"year":2025,"cited_by_count":3}],"updated_date":"2026-05-03T08:25:01.440150","created_date":"2025-11-28T00:00:00"}