{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004602918,0.0001343468,0.0001407502,0.0002062774,0.0007830877,0.0002913242,0.004974325,0.00003153866,0.000003639135],"category_scores_gemma":[0.0004073672,0.0001144724,0.00001314275,0.001042104,0.00002106978,0.001072018,0.004119829,0.0009433426,0.000006177479],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000115494,"about_ca_system_score_gemma":0.0001365253,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0007289162,"about_ca_topic_score_gemma":0.001897091,"domain_scores_codex":[0.9973632,0.0005872233,0.0004454954,0.0008442729,0.0004375828,0.000322216],"domain_scores_gemma":[0.9966898,0.000265991,0.0001962578,0.002778824,0.000027957,0.00004114688],"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":[0.000006753176,0.0001244542,0.0184397,0.00001001707,0.000009115655,0.000004584055,0.001798788,0.003208269,0.0007363058,0.01019218,0.000569706,0.9649001],"study_design_scores_gemma":[0.0001995893,0.000024993,0.004779937,0.00001052686,0.000002323961,0.000004231888,0.005518199,0.8500055,0.00001749732,0.0002976833,0.138983,0.0001564575],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01324878,0.0005307669,0.9608479,0.01811765,0.000141447,0.001547926,0.0003040285,0.0009705073,0.004290968],"genre_scores_gemma":[0.958504,0.0001263593,0.03628683,0.000249762,0.00004284433,0.0002924131,0.004236145,0.00001458657,0.0002471052],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9647437,"threshold_uncertainty_score":0.9243616,"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."}}