{"id":"W2964212410","doi":"","title":"On calibration of modern neural networks","year":2017,"lang":"en","type":"article","venue":"International Conference on Machine Learning","topic":"Anomaly Detection Techniques and Applications","field":"Computer Science","cited_by":1185,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Normalization (sociology); Computer science; Artificial neural network; Correctness; Calibration; Scaling; Deep neural networks; Artificial intelligence; Machine learning; Simple (philosophy); Deep learning; Pattern recognition (psychology); Data mining; Algorithm; Mathematics; Statistics","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.000112065,0.00008732355,0.00008597664,0.00007825252,0.0002678724,0.0002464548,0.0009363792,0.00004351353,0.00007741734],"category_scores_gemma":[0.0000597059,0.00008086432,0.0000492567,0.00003355946,0.0000385169,0.0002955308,0.0001487578,0.0002706188,0.000006331112],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001898098,"about_ca_system_score_gemma":0.00001336501,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006325945,"about_ca_topic_score_gemma":0.000009126119,"domain_scores_codex":[0.9992843,0.00003303991,0.0001583918,0.0002228696,0.000211185,0.00009026213],"domain_scores_gemma":[0.9991807,0.00004273229,0.0002688504,0.0003714435,0.0001020917,0.0000341598],"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":[0.00001625396,0.00003500897,0.001591315,0.000001230505,0.0000091963,0.00000159769,0.00003813208,0.0375682,0.0006951758,0.8622808,0.00003506221,0.09772804],"study_design_scores_gemma":[0.0001137879,0.0001195351,0.002620645,0.00001632098,0.000001152063,0.000002365043,0.000002183393,0.9769416,0.0008270384,0.01907802,0.0002029623,0.00007438334],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01713523,0.000005412916,0.9618363,0.002525287,0.0001728534,0.00006941536,0.000002550612,0.0001207393,0.0181322],"genre_scores_gemma":[0.9968289,0.00001221382,0.0022181,0.0001655951,0.00005570887,0.00001596075,0.000008767826,0.00000602589,0.0006887113],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9796937,"threshold_uncertainty_score":0.3297551,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03526583321291214,"score_gpt":0.3120534779422832,"score_spread":0.276787644729371,"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."}}