{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0008681192,0.0001179188,0.0001576705,0.0001729679,0.0001569592,0.0000581225,0.004966399,0.00001695986,0.0001332201],"category_scores_gemma":[0.0002274941,0.0001040813,0.00004749417,0.0003470493,0.00004433669,0.0005922347,0.005497443,0.00037283,0.000001578712],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003758943,"about_ca_system_score_gemma":0.0000346035,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003159336,"about_ca_topic_score_gemma":0.000001619123,"domain_scores_codex":[0.9981307,0.00002830248,0.0004255594,0.0004101492,0.0008805349,0.0001247609],"domain_scores_gemma":[0.9982321,0.0001111272,0.0009228186,0.0004820549,0.000230042,0.00002184209],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00005609847,0.0003424824,0.001718706,0.0002002529,0.0001772161,5.009333e-7,0.0002373316,0.06650443,0.004790291,0.9024962,0.001149485,0.02232707],"study_design_scores_gemma":[0.000261154,0.00005920517,0.0007473597,0.0001092361,0.0000226823,0.000001059958,0.000502049,0.989136,0.0003758832,0.006795807,0.001895671,0.00009391975],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1884836,0.0004051517,0.6243018,0.06258444,0.001852498,0.001908151,0.007458113,0.0004570449,0.1125492],"genre_scores_gemma":[0.9734048,0.0001817604,0.02429051,0.0001147447,0.00001820763,0.0000112777,0.000564348,0.000009288126,0.001405084],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9226316,"threshold_uncertainty_score":0.9228888,"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."}}