{"id":"W3081806177","doi":"10.1109/ccwc51732.2021.9376126","title":"Reinforcement Learning Based Penetration Testing of a Microgrid Control Algorithm","year":2021,"lang":"en","type":"preprint","venue":"","topic":"Smart Grid Security and Resilience","field":"Engineering","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal","funders":"","keywords":"Reinforcement learning; Computer science; Relay; Microgrid; Adversary; Controller (irrigation); Distributed computing; Software; Control (management); Electric power system; Embedded system; Computer security; Artificial intelligence; Power (physics); Operating system","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001300378,0.0008010403,0.0006390714,0.000324391,0.0002713882,0.000560873,0.0008811812,0.0009990466,0.001850611],"category_scores_gemma":[0.00592443,0.0002381243,0.0002758603,0.0001265193,0.0009946873,0.0005381817,0.0006891171,0.0008741495,0.0001774595],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006910695,"about_ca_system_score_gemma":0.0009014201,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005756135,"about_ca_topic_score_gemma":0.003328424,"domain_scores_codex":[0.9995736,0.0001588601,0.00002168675,0.00007981037,0.00007600924,0.0000899095],"domain_scores_gemma":[0.9965277,0.002434044,0.0003311779,0.0001707762,0.0003745343,0.000161658],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001297857,0.0001076965,0.001525469,0.00004277193,0.00002117974,0.00009875751,0.00006472149,0.983017,0.001454635,0.001169149,0.0002318916,0.01213698],"study_design_scores_gemma":[0.00001282135,0.00007618784,0.0001135412,0.000004438943,0.000003437215,0.000009488275,0.000007521363,0.9989495,0.0004509041,0.0003072465,0.00006249596,0.000002461328],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6877493,0.0003037691,0.2981055,0.0008600875,0.0000718134,0.0003190159,0.00006607047,0.001317696,0.01120671],"genre_scores_gemma":[0.9901125,0.0000181193,0.009129044,0.00005162201,0.000003081984,0.00004659298,0.00001834936,0.00001082967,0.0006099291],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.005756135,"threshold_uncertainty_score":0.01144528,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01098435493986431,"score_gpt":0.2077113830321464,"score_spread":0.1967270280922821,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}