{"id":"W4288048294","doi":"10.1109/dsn-w54100.2022.00020","title":"Towards Building Resilient Ensembles against Training Data Faults","year":2022,"lang":"en","type":"article","venue":"","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Computer science; Construct (python library); Training (meteorology); Training set; Fault injection; Machine learning; Data modeling; Artificial intelligence; Distributed computing; Data mining; Software engineering; Computer network; Software","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":[0.003231778,0.001451915,0.001310115,0.001082574,0.0008117298,0.001397095,0.001803701,0.001779124,0.002298194],"category_scores_gemma":[0.01283134,0.001105637,0.001187692,0.0005098135,0.001473396,0.002672246,0.004043032,0.004087041,0.001819098],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006647891,"about_ca_system_score_gemma":0.0008292284,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006819688,"about_ca_topic_score_gemma":0.0008196356,"domain_scores_codex":[0.9979934,0.0005728015,0.00009741887,0.0003715811,0.0007293471,0.0002353887],"domain_scores_gemma":[0.9948859,0.0018932,0.0003628944,0.00171161,0.0009374728,0.0002088404],"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":[0.00007826465,0.00007260766,0.001205916,0.0000499659,0.000113337,0.0001102699,0.0001457101,0.9160963,0.01725909,0.01200378,0.001661156,0.05120358],"study_design_scores_gemma":[0.000004548493,0.00005335545,0.0001377478,0.00001188146,0.00001594367,0.00004379993,0.00001785324,0.9736626,0.007764653,0.01718003,0.001094779,0.00001287312],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01656307,0.00008447016,0.9806574,0.0001850104,0.00004848489,0.00004177817,0.00005768484,0.001385901,0.000976272],"genre_scores_gemma":[0.5896145,0.0002271758,0.4050449,0.000685464,0.0001670659,0.0002758777,0.0005157898,0.0006447846,0.002824383],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003231778,"threshold_uncertainty_score":0.01709151,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06867818540571483,"score_gpt":0.3158087203122173,"score_spread":0.2471305349065025,"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."}}