{"id":"W4320717670","doi":"10.1016/j.knosys.2023.110384","title":"Label noise analysis meets adversarial training: A defense against label poisoning in federated learning","year":2023,"lang":"en","type":"article","venue":"Knowledge-Based Systems","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":36,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of New Brunswick; University of Windsor","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Backdoor; Computer science; Noise (video); Adversarial system; Artificial intelligence; Machine learning; Artificial noise; The Internet; Differential privacy; Trojan; Data mining; Key (lock); Computer security; Computer network; World Wide Web","routes":{"ca_aff":true,"ca_fund":true,"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":[0.009309467,0.001361461,0.002127602,0.001281711,0.001745327,0.002525182,0.003564641,0.005312326,0.001158213],"category_scores_gemma":[0.03523502,0.000849654,0.00123877,0.001077098,0.004510747,0.004908845,0.008278188,0.006018049,0.00058235],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001738264,"about_ca_system_score_gemma":0.002087501,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001315531,"about_ca_topic_score_gemma":0.001044556,"domain_scores_codex":[0.9923955,0.002968403,0.0002702245,0.001418363,0.002366737,0.0005808097],"domain_scores_gemma":[0.9699358,0.01683344,0.001991062,0.008591408,0.002128844,0.0005194059],"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":[0.001573956,0.0004689954,0.004095978,0.0002732651,0.0003577521,0.0004881066,0.0006070702,0.5859758,0.02094549,0.1255956,0.009753394,0.2498647],"study_design_scores_gemma":[0.00001668363,0.00005826587,0.0001703565,0.000018411,0.00001974673,0.00009773716,0.00002620804,0.9386483,0.005168267,0.05518721,0.0005750822,0.00001370646],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01978375,0.0002430985,0.9763326,0.0009683924,0.00007954545,0.00005892192,0.00005960567,0.001085793,0.001388293],"genre_scores_gemma":[0.842241,0.0001531622,0.1534918,0.0009092636,0.000147801,0.0001028235,0.0001600034,0.0001787883,0.002615352],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.009309467,"threshold_uncertainty_score":0.04923379,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04187831450664831,"score_gpt":0.2963231359441755,"score_spread":0.2544448214375272,"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."}}