{"id":"W4287883233","doi":"10.1109/dsn53405.2022.00027","title":"The Fault in Our Data Stars: Studying Mitigation Techniques against Faulty Training Data in Machine Learning Applications","year":2022,"lang":"en","type":"article","venue":"","topic":"Machine Learning and Data Classification","field":"Computer Science","cited_by":15,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia; University of British Columbia Hospital","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Smoothing; Machine learning; Training (meteorology); Training set; Artificial intelligence; Resilience (materials science); Quality (philosophy); Fault tolerance; Data modeling; Data mining; Distributed computing; Database","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":[0.01078344,0.001189769,0.0009270126,0.002677726,0.001083838,0.001346282,0.001864129,0.001881039,0.0004786286],"category_scores_gemma":[0.05863264,0.0003489346,0.0009462932,0.002126581,0.001268754,0.004071282,0.001702997,0.001962723,0.0004256674],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006554705,"about_ca_system_score_gemma":0.001202419,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00146088,"about_ca_topic_score_gemma":0.001438481,"domain_scores_codex":[0.992864,0.00318338,0.0005673874,0.001203833,0.001811515,0.0003698271],"domain_scores_gemma":[0.9302537,0.04681054,0.004714669,0.01065995,0.006934498,0.0006265833],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001266301,0.0006466542,0.06201452,0.0022955,0.0007070884,0.0004044641,0.001790134,0.1204437,0.01828181,0.005710833,0.008792588,0.7776465],"study_design_scores_gemma":[0.000137061,0.004262446,0.04976469,0.001281947,0.001130034,0.001805075,0.002935895,0.7689065,0.1093272,0.03330807,0.02691312,0.0002281127],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6047804,0.0269151,0.3546096,0.003837873,0.0009090085,0.0003561443,0.0007396477,0.00414205,0.003710228],"genre_scores_gemma":[0.8886437,0.002975824,0.1054642,0.0004960094,0.000270102,0.0001152236,0.000895814,0.0002468334,0.0008922883],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01078344,"threshold_uncertainty_score":0.05702901,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08916431159158557,"score_gpt":0.3410178916331216,"score_spread":0.2518535800415361,"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."}}