{"id":"W3039588859","doi":"10.32920/24132885","title":"Towards Robust Deep Learning With Ensemble Networks and Noisy Layers","year":2023,"lang":"en","type":"preprint","venue":"","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Toronto Metropolitan University; Vector Institute; McMaster University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Robustness (evolution); Adversarial system; Computer science; Artificial intelligence; Deep learning; Machine learning","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","research_integrity"],"consensus_categories":[],"category_scores_codex":[0.0008189943,0.0005062143,0.0005420251,0.0002386479,0.0003513962,0.0007077762,0.001396293,0.0004202042,0.00002584684],"category_scores_gemma":[0.0001834183,0.0004332944,0.00009521664,0.0004770272,0.0001087689,0.0003160494,0.005200638,0.002511407,0.00003388795],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001025041,"about_ca_system_score_gemma":0.0001587246,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006607425,"about_ca_topic_score_gemma":0.0002396641,"domain_scores_codex":[0.996897,0.0002477737,0.0003463452,0.001322983,0.0005328843,0.0006529629],"domain_scores_gemma":[0.9981805,0.0002830956,0.0002965532,0.0008987007,0.0001350102,0.0002061474],"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":[0.00001150506,0.000008250797,0.002466799,0.00005357365,0.00007171097,0.0000942346,0.0004960819,0.958375,0.000001766925,0.005605198,0.0001366892,0.03267922],"study_design_scores_gemma":[0.0003106183,0.00008073877,0.002656775,0.0001446848,0.00003200061,0.00002981887,0.0001471171,0.9943432,0.000007664338,0.001144677,0.0005187516,0.0005840156],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.002040237,0.0002026792,0.9868794,0.001033215,0.00101393,0.0002950121,3.46194e-7,0.001536703,0.006998486],"genre_scores_gemma":[0.567912,0.0002113287,0.4270996,0.0002255996,0.0004700049,0.00005552332,0.0000235562,0.0001198279,0.003882492],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.5658718,"threshold_uncertainty_score":0.9998119,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02459965081981148,"score_gpt":0.2520518520576505,"score_spread":0.227452201237839,"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."}}