{"id":"W3120768806","doi":"10.48550/arxiv.2101.05036","title":"Estimating and Evaluating Regression Predictive Uncertainty in Deep Object Detectors","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Artificial intelligence; Machine learning; Correctness; Probabilistic logic; Minimum bounding box; Variance (accounting); Entropy (arrow of time); Regression; Scoring rule; Data mining; Statistics; Algorithm; Mathematics","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0009258903,0.0003856744,0.0004661496,0.0003482898,0.0002646227,0.0002059443,0.001100451,0.0003366026,0.00001501538],"category_scores_gemma":[0.001012997,0.000427963,0.0001180377,0.0008085556,0.0001150948,0.0005772532,0.004288846,0.001423642,0.000002671401],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004885737,"about_ca_system_score_gemma":0.0003253364,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004971132,"about_ca_topic_score_gemma":0.0001675474,"domain_scores_codex":[0.9968114,0.0007459778,0.0002994433,0.001531179,0.0001939207,0.0004180523],"domain_scores_gemma":[0.9978456,0.0005352272,0.000433613,0.0008719429,0.0001758252,0.0001377745],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00002665667,0.00002212577,0.01173987,0.00009200168,0.00003257031,0.0003612523,0.0016764,0.9785016,0.0000402535,0.001155325,0.00000125227,0.006350696],"study_design_scores_gemma":[0.0005643638,0.00006255787,0.005006828,0.0007785437,0.000043279,0.000009031287,0.0005434706,0.9856867,0.00002988316,0.006885241,9.621874e-7,0.0003891227],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4792325,0.00008731241,0.5197232,0.00001952544,0.0004354688,0.0001789215,9.359526e-7,0.0001239916,0.0001980961],"genre_scores_gemma":[0.9422451,0.00002663075,0.0575438,0.00002314244,0.00008176115,0.000002019762,0.00001022091,0.00002158839,0.00004578094],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.4630125,"threshold_uncertainty_score":0.9998172,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05686327775697341,"score_gpt":0.2433853414853592,"score_spread":0.1865220637283858,"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."}}