{"id":"https://openalex.org/W4417299963","doi":"https://doi.org/10.1016/j.jisa.2026.104507","title":"Unveiling the black box: A multi-layer framework for explaining reinforcement learning-based cyber agents","display_name":"Unveiling the black box: A multi-layer framework for explaining reinforcement learning-based cyber agents","publication_year":2026,"publication_date":"2026-06-02","ids":{"openalex":"https://openalex.org/W4417299963","doi":"https://doi.org/10.1016/j.jisa.2026.104507"},"language":"en","primary_location":{"id":"doi:10.1016/j.jisa.2026.104507","is_oa":true,"landing_page_url":"https://doi.org/10.1016/j.jisa.2026.104507","pdf_url":null,"source":{"id":"https://openalex.org/S4210191536","display_name":"Journal of Information Security and Applications","issn_l":"2214-2126","issn":["2214-2126","2214-2134"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320990","host_organization_name":"Elsevier BV","host_organization_lineage":["https://openalex.org/P4310320990"],"host_organization_lineage_names":["Elsevier BV"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Journal of Information Security and Applications","raw_type":"journal-article"},"type":"article","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://doi.org/10.1016/j.jisa.2026.104507","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5084412291","display_name":"Diksha Goel","orcid":"https://orcid.org/0000-0001-8212-8793"},"institutions":[{"id":"https://openalex.org/I4210154771","display_name":"CSIRO Manufacturing","ror":"https://ror.org/04sx9wp33","country_code":"AU","type":"facility","lineage":["https://openalex.org/I1292875679","https://openalex.org/I2801453606","https://openalex.org/I4210154771","https://openalex.org/I4387156119"]},{"id":"https://openalex.org/I42894916","display_name":"Data61","ror":"https://ror.org/03q397159","country_code":"AU","type":"other","lineage":["https://openalex.org/I1292875679","https://openalex.org/I2801453606","https://openalex.org/I42894916","https://openalex.org/I4387156119"]}],"countries":["AU"],"is_corresponding":true,"raw_author_name":"Diksha Goel","raw_affiliation_strings":["CSIRO\u2019s Data61, Clayton, Australia","Cyber Security Cooperative Research Centre (CSCRC), Joondalup, Australia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"CSIRO\u2019s Data61, Clayton, Australia","institution_ids":["https://openalex.org/I4210154771","https://openalex.org/I42894916"]},{"raw_affiliation_string":"Cyber Security Cooperative Research Centre (CSCRC), Joondalup, Australia","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5037106850","display_name":"Kristen Moore","orcid":"https://orcid.org/0000-0002-9962-5080"},"institutions":[{"id":"https://openalex.org/I4210154771","display_name":"CSIRO Manufacturing","ror":"https://ror.org/04sx9wp33","country_code":"AU","type":"facility","lineage":["https://openalex.org/I1292875679","https://openalex.org/I2801453606","https://openalex.org/I4210154771","https://openalex.org/I4387156119"]},{"id":"https://openalex.org/I42894916","display_name":"Data61","ror":"https://ror.org/03q397159","country_code":"AU","type":"other","lineage":["https://openalex.org/I1292875679","https://openalex.org/I2801453606","https://openalex.org/I42894916","https://openalex.org/I4387156119"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Kristen Moore","raw_affiliation_strings":["CSIRO\u2019s Data61, Clayton, Australia","Cyber Security Cooperative Research Centre (CSCRC), Joondalup, Australia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"CSIRO\u2019s Data61, Clayton, Australia","institution_ids":["https://openalex.org/I4210154771","https://openalex.org/I42894916"]},{"raw_affiliation_string":"Cyber Security Cooperative Research Centre (CSCRC), Joondalup, Australia","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109712935","display_name":"Jeff Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I4210154771","display_name":"CSIRO Manufacturing","ror":"https://ror.org/04sx9wp33","country_code":"AU","type":"facility","lineage":["https://openalex.org/I1292875679","https://openalex.org/I2801453606","https://openalex.org/I4210154771","https://openalex.org/I4387156119"]},{"id":"https://openalex.org/I42894916","display_name":"Data61","ror":"https://ror.org/03q397159","country_code":"AU","type":"other","lineage":["https://openalex.org/I1292875679","https://openalex.org/I2801453606","https://openalex.org/I42894916","https://openalex.org/I4387156119"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Jeff Wang","raw_affiliation_strings":["CSIRO\u2019s Data61, Clayton, Australia","Cyber Security Cooperative Research Centre (CSCRC), Joondalup, Australia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"CSIRO\u2019s Data61, Clayton, Australia","institution_ids":["https://openalex.org/I4210154771","https://openalex.org/I42894916"]},{"raw_affiliation_string":"Cyber Security Cooperative Research Centre (CSCRC), Joondalup, Australia","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048500989","display_name":"Minjune Kim","orcid":"https://orcid.org/0000-0002-6161-7982"},"institutions":[{"id":"https://openalex.org/I4210154771","display_name":"CSIRO Manufacturing","ror":"https://ror.org/04sx9wp33","country_code":"AU","type":"facility","lineage":["https://openalex.org/I1292875679","https://openalex.org/I2801453606","https://openalex.org/I4210154771","https://openalex.org/I4387156119"]},{"id":"https://openalex.org/I42894916","display_name":"Data61","ror":"https://ror.org/03q397159","country_code":"AU","type":"other","lineage":["https://openalex.org/I1292875679","https://openalex.org/I2801453606","https://openalex.org/I42894916","https://openalex.org/I4387156119"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Minjune Kim","raw_affiliation_strings":["CSIRO\u2019s Data61, Clayton, Australia","Cyber Security Cooperative Research Centre (CSCRC), Joondalup, Australia"],"raw_orcid":"https://orcid.org/0000-0002-6161-7982","affiliations":[{"raw_affiliation_string":"CSIRO\u2019s Data61, Clayton, Australia","institution_ids":["https://openalex.org/I4210154771","https://openalex.org/I42894916"]},{"raw_affiliation_string":"Cyber Security Cooperative Research Centre (CSCRC), Joondalup, Australia","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5085593383","display_name":"Thanh Thi Nguyen","orcid":"https://orcid.org/0000-0001-9709-1663"},"institutions":[{"id":"https://openalex.org/I56590836","display_name":"Monash University","ror":"https://ror.org/02bfwt286","country_code":"AU","type":"education","lineage":["https://openalex.org/I56590836"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Thanh Thi Nguyen","raw_affiliation_strings":["Monash University, Melbourne, Australia"],"raw_orcid":"https://orcid.org/0000-0001-9709-1663","affiliations":[{"raw_affiliation_string":"Monash University, Melbourne, Australia","institution_ids":["https://openalex.org/I56590836"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5084412291"],"corresponding_institution_ids":["https://openalex.org/I4210154771","https://openalex.org/I42894916"],"apc_list":{"value":2860,"currency":"USD","value_usd":2860},"apc_paid":{"value":2860,"currency":"USD","value_usd":2860},"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.0206705,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"101","issue":null,"first_page":"104507","last_page":"104507"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.40049999952316284,"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":0.40049999952316284,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.2903999984264374,"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/T10734","display_name":"Information and Cyber Security","score":0.052799999713897705,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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":0.7967000007629395},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.6269000172615051},{"id":"https://openalex.org/keywords/adversarial-system","display_name":"Adversarial system","score":0.6151000261306763},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.6067000031471252},{"id":"https://openalex.org/keywords/markov-decision-process","display_name":"Markov decision process","score":0.5978000164031982},{"id":"https://openalex.org/keywords/autonomous-agent","display_name":"Autonomous agent","score":0.37929999828338623},{"id":"https://openalex.org/keywords/partially-observable-markov-decision-process","display_name":"Partially observable Markov decision process","score":0.3716000020503998}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.7967000007629395},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.6269000172615051},{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.6151000261306763},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.6067000031471252},{"id":"https://openalex.org/C106189395","wikidata":"https://www.wikidata.org/wiki/Q176789","display_name":"Markov decision process","level":3,"score":0.5978000164031982},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5902000069618225},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.40700000524520874},{"id":"https://openalex.org/C13687954","wikidata":"https://www.wikidata.org/wiki/Q4826847","display_name":"Autonomous agent","level":2,"score":0.37929999828338623},{"id":"https://openalex.org/C17098449","wikidata":"https://www.wikidata.org/wiki/Q176814","display_name":"Partially observable Markov decision process","level":4,"score":0.3716000020503998},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.3603000044822693},{"id":"https://openalex.org/C2984634286","wikidata":"https://www.wikidata.org/wiki/Q1331926","display_name":"Decision process","level":2,"score":0.34119999408721924},{"id":"https://openalex.org/C67203356","wikidata":"https://www.wikidata.org/wiki/Q1321905","display_name":"Reinforcement","level":2,"score":0.32690000534057617},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.3082999885082245},{"id":"https://openalex.org/C108170787","wikidata":"https://www.wikidata.org/wiki/Q3951828","display_name":"Agency (philosophy)","level":2,"score":0.29260000586509705},{"id":"https://openalex.org/C94966114","wikidata":"https://www.wikidata.org/wiki/Q29256","display_name":"Black box","level":2,"score":0.2888999879360199},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.2856999933719635},{"id":"https://openalex.org/C159886148","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov process","level":2,"score":0.26980000734329224},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.2696000039577484}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1016/j.jisa.2026.104507","is_oa":true,"landing_page_url":"https://doi.org/10.1016/j.jisa.2026.104507","pdf_url":null,"source":{"id":"https://openalex.org/S4210191536","display_name":"Journal of Information Security and Applications","issn_l":"2214-2126","issn":["2214-2126","2214-2134"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320990","host_organization_name":"Elsevier BV","host_organization_lineage":["https://openalex.org/P4310320990"],"host_organization_lineage_names":["Elsevier BV"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Journal of Information Security and Applications","raw_type":"journal-article"},{"id":"pmh:oai:arXiv.org:2505.11708","is_oa":true,"landing_page_url":"https://arxiv.org/abs/2505.11708","pdf_url":"https://arxiv.org/pdf/2505.11708","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,"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.2505.11708","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2505.11708","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,"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":"Preprint"}],"best_oa_location":{"id":"doi:10.1016/j.jisa.2026.104507","is_oa":true,"landing_page_url":"https://doi.org/10.1016/j.jisa.2026.104507","pdf_url":null,"source":{"id":"https://openalex.org/S4210191536","display_name":"Journal of Information Security and Applications","issn_l":"2214-2126","issn":["2214-2126","2214-2134"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320990","host_organization_name":"Elsevier BV","host_organization_lineage":["https://openalex.org/P4310320990"],"host_organization_lineage_names":["Elsevier BV"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Journal of Information Security and Applications","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Reinforcement":[0],"Learning":[1],"(RL)":[2],"agents":[3,54],"are":[4,35,137],"increasingly":[5],"used":[6],"to":[7,81,105,189],"simulate":[8],"sophisticated":[9],"cyberattacks,":[10],"but":[11],"their":[12],"decision-making":[13],"processes":[14],"remain":[15],"opaque,":[16],"hindering":[17],"trust,":[18],"debugging,":[19,161],"and":[20,37,63,85,99,110,150,165,192,196],"defensive":[21],"preparedness.":[22],"In":[23,41],"high-stakes":[24],"cybersecurity":[25],"contexts,":[26],"explainability":[27,49],"is":[28,147],"essential":[29],"for":[30,51],"understanding":[31,197],"how":[32],"adversarial":[33],"strategies":[34],"formed":[36],"evolve":[38],"over":[39],"time.":[40],"this":[42],"paper,":[43],"we":[44,70,92],"propose":[45],"a":[46,74],"unified,":[47],"multi-layer":[48],"framework":[50,122,178],"RL-based":[52],"attacker":[53],"that":[55],"reveals":[56],"both":[57,148],"strategic":[58],"(Markov":[59],"Decision":[60,78],"Process":[61,79],"(MDP)-level)":[62],"tactical":[64],"(policy-level)":[65],"reasoning.":[66],"At":[67,89],"the":[68,90,94,187],"MDP-level,":[69],"model":[71],"cyberattacks":[72],"as":[73,156],"Partially":[75],"Observable":[76],"Markov":[77],"(POMDP)":[80],"expose":[82],"exploration-exploitation":[83],"dynamics":[84],"phase-aware":[86,162],"behavioural":[87,175],"shifts.":[88],"policy-level,":[91],"analyse":[93],"temporal":[95],"evolution":[96],"of":[97,118],"Q-values":[98],"use":[100,153],"Prioritised":[101],"Experience":[102],"Replay":[103],"(PER)":[104],"surface":[106],"critical":[107],"learning":[108,172,184],"transitions":[109],"evolving":[111],"action":[112],"preferences.":[113],"Evaluated":[114],"across":[115],"CyberBattleSim":[116],"environments":[117],"increasing":[119],"complexity,":[120],"our":[121,145,177],"offers":[123],"interpretable":[124,174],"insights":[125],"into":[126,173,182],"agent":[127,183],"behaviour":[128],"at":[129],"scale.":[130],"Unlike":[131],"previous":[132],"explainable":[133],"RL":[134,159],"methods,":[135],"which":[136],"predominantly":[138],"post-hoc,":[139],"domain-specific,":[140],"or":[141],"limited":[142],"in":[143,194],"depth,":[144],"approach":[146],"agent-":[149],"environment-agnostic,":[151],"supporting":[152],"cases":[154],"such":[155],"red-team":[157],"simulation,":[158],"policy":[160],"threat":[163],"modelling":[164],"anticipatory":[166],"defence":[167],"planning.":[168],"By":[169],"transforming":[170],"black-box":[171],"signals,":[176],"provides":[179],"diagnostic":[180],"insight":[181],"dynamics,":[185],"with":[186],"potential":[188],"support":[190],"developers":[191],"defenders":[193],"analysing":[195],"autonomous":[198],"cyber":[199],"threats.":[200]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2025-10-10T00:00:00"}
