{"id":"W4281788841","doi":"10.1145/3514221.3517904","title":"FILA: Online Auditing of Machine Learning Model Accuracy under Finite Labelling Budget","year":2022,"lang":"en","type":"article","venue":"Proceedings of the 2022 International Conference on Management of Data","topic":"Machine Learning and Data Classification","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Audit; Pipeline (software); Computer science; Process (computing); Training set; Artificial intelligence; Machine learning; Data modeling; Labelling; Software engineering; Accounting; Programming language","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":[0.01463815,0.002163974,0.002434143,0.003297581,0.001722676,0.004373042,0.006299646,0.003393457,0.0106478],"category_scores_gemma":[0.09741161,0.001158696,0.00115806,0.001927853,0.002192554,0.007761076,0.007345913,0.00505098,0.006044118],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002977163,"about_ca_system_score_gemma":0.005640576,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007262623,"about_ca_topic_score_gemma":0.01036108,"domain_scores_codex":[0.9877465,0.004396557,0.0008211723,0.002043171,0.003907669,0.001084824],"domain_scores_gemma":[0.9276262,0.0324096,0.004239812,0.02572016,0.007967071,0.002037121],"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":[0.005669194,0.001279251,0.03571964,0.001099081,0.0006353305,0.0005961398,0.0004791223,0.2258074,0.0158316,0.02244583,0.1186925,0.5717449],"study_design_scores_gemma":[0.0001324897,0.0002205942,0.002372875,0.00005532894,0.00002928398,0.0001095179,0.00006729644,0.9710727,0.007990278,0.01405052,0.003850145,0.00004896961],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1157774,0.001620091,0.7223246,0.002349356,0.0009840757,0.0007756338,0.008537092,0.1366549,0.01097678],"genre_scores_gemma":[0.7682915,0.0003087303,0.2064679,0.0007047969,0.0003197684,0.0007116502,0.01368784,0.003971652,0.005536242],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01463815,"threshold_uncertainty_score":0.07741487,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1060019905566788,"score_gpt":0.3295175570380833,"score_spread":0.2235155664814045,"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."}}