{"id":"W4323556281","doi":"10.48550/arxiv.2303.02322","title":"Improved Robustness Against Adaptive Attacks With Ensembles and Error-Correcting Output Codes","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Robustness (evolution); Computer science; Machine learning; Artificial intelligence; Convolutional neural network","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002151902,0.000904486,0.0007273073,0.0006176243,0.0005857476,0.0009570126,0.0009895298,0.001182439,0.001115879],"category_scores_gemma":[0.01176994,0.0002813817,0.0005634272,0.0005431818,0.001392384,0.001586454,0.002276271,0.002061196,0.0003809822],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005650118,"about_ca_system_score_gemma":0.0005984009,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0008311175,"about_ca_topic_score_gemma":0.000869198,"domain_scores_codex":[0.9980332,0.0005255204,0.0001011561,0.0003351122,0.0007446785,0.0002603313],"domain_scores_gemma":[0.992727,0.003306123,0.0006981405,0.002093291,0.0009991814,0.0001763345],"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":[0.0001559219,0.00004036212,0.00259622,0.00007195435,0.0001273051,0.0001421622,0.00009249623,0.8870363,0.02150621,0.03642193,0.001007269,0.05080189],"study_design_scores_gemma":[0.000006994102,0.00009847658,0.0004580558,0.00001972185,0.00002226217,0.0001157674,0.00002191703,0.9655246,0.01546933,0.01706521,0.001177409,0.00002037374],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1533117,0.0006411732,0.8378335,0.0005355609,0.0001592215,0.00005739538,0.000116545,0.0007740294,0.006570951],"genre_scores_gemma":[0.9475024,0.0002764249,0.05031767,0.0001305187,0.00005892136,0.00004377219,0.00009231555,0.00006976096,0.001508241],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.002151902,"threshold_uncertainty_score":0.01138049,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08886710133815569,"score_gpt":0.2166347461521532,"score_spread":0.1277676448139975,"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."}}