{"id":"W4389722721","doi":"10.1109/tsg.2023.3343100","title":"On the Use of Reinforcement Learning for Attacking and Defending Load Frequency Control","year":2023,"lang":"en","type":"article","venue":"IEEE Transactions on Smart Grid","topic":"Smart Grid Security and Resilience","field":"Engineering","cited_by":23,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Reinforcement learning; Computer science; Sophistication; Anomaly detection; Computer security; Grid; Process (computing); Reliability (semiconductor); Control (management); Artificial intelligence; Power (physics)","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":[0.001774905,0.0006496537,0.0004881821,0.0003608714,0.0002615713,0.0006871599,0.0008349809,0.0006937523,0.001308533],"category_scores_gemma":[0.008113849,0.0002918797,0.0003335086,0.0001883162,0.001111348,0.000950782,0.0007765501,0.001343729,0.0001987369],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008991727,"about_ca_system_score_gemma":0.0009929754,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004565327,"about_ca_topic_score_gemma":0.004171506,"domain_scores_codex":[0.9992925,0.0003318539,0.00003038342,0.0001310296,0.0001429599,0.0000712464],"domain_scores_gemma":[0.9957566,0.003167124,0.000290576,0.0002559672,0.0004027603,0.0001269556],"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":[0.00007535479,0.00006835668,0.001538639,0.00003217303,0.00003204143,0.000044298,0.00004055588,0.9401062,0.002365523,0.01177671,0.0004058795,0.04351433],"study_design_scores_gemma":[0.000002967781,0.00001787162,0.00005790003,0.000002806055,0.000002157952,0.000004694612,0.000001351058,0.9969715,0.0004092549,0.002423683,0.0001037024,0.000002122102],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03664743,0.0002161703,0.9586976,0.0005212884,0.0000356607,0.00003914665,0.00002273751,0.0005374562,0.003282537],"genre_scores_gemma":[0.8993837,0.0001443651,0.09834144,0.0002060738,0.00003459779,0.00005747146,0.00003494195,0.00006132467,0.001735944],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004565327,"threshold_uncertainty_score":0.009386718,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03901159060105528,"score_gpt":0.2383037907494854,"score_spread":0.1992922001484301,"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."}}