{"id":"W4413318932","doi":"10.1109/jiot.2025.3599106","title":"Securing Smart Grid Federated Learning Against Advanced Evasion Attacks Using Ensemble-Based Adversarial Training","year":2025,"lang":"en","type":"article","venue":"IEEE Internet of Things Journal","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":false,"ca_institutions":"Queen's University","funders":"National Science Foundation","keywords":"Adversarial system; Computer science; Training (meteorology); Evasion (ethics); Smart grid; Ensemble learning; Artificial intelligence; Grid; Computer security; Machine learning; Training set; Computer network; Engineering","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.001814343,0.0004006138,0.000608521,0.0006762561,0.0006056362,0.0005865401,0.001406968,0.0002179874,0.00002155106],"category_scores_gemma":[0.0008878781,0.000406667,0.0003241754,0.0006351336,0.00009867869,0.001711893,0.0003638469,0.002136622,0.000005577434],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004250683,"about_ca_system_score_gemma":0.0005657088,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007091324,"about_ca_topic_score_gemma":0.000004936931,"domain_scores_codex":[0.996484,0.0005068842,0.00103419,0.0005815825,0.000728045,0.0006652407],"domain_scores_gemma":[0.9975835,0.0004461523,0.001059641,0.0003146869,0.0004201675,0.0001758391],"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":[0.0003303982,0.00006593165,0.001366202,0.00008078868,0.0001741359,0.0001717126,0.005115276,0.8403461,0.07640273,0.0003896009,0.0001853276,0.07537182],"study_design_scores_gemma":[0.00249763,0.00016694,0.00009032391,0.001757849,0.00004397284,0.00009781703,0.0003897887,0.9557408,0.03748493,0.0003378103,0.001016514,0.0003756311],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3501134,0.00007307721,0.6434159,0.0002240215,0.004620747,0.0001086586,2.929847e-7,0.0001438119,0.001300083],"genre_scores_gemma":[0.9022072,0.000009614719,0.0967522,0.0004446587,0.0003381096,0.000002343104,0.00000269647,0.00003508354,0.0002080973],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.5520938,"threshold_uncertainty_score":0.9998385,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02371606787940691,"score_gpt":0.2903333922325174,"score_spread":0.2666173243531105,"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."}}