{"id":"W3206981139","doi":"10.5592/co/cetra.2020.1218","title":"Uncertainty estimation on road safety analysis using bayesian deep neural networks","year":2021,"lang":"en","type":"article","venue":"Road and rail infrastructure","topic":"Traffic Prediction and Management Techniques","field":"Engineering","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Artificial neural network; Deep learning; Bayesian probability; Artificial intelligence; Machine learning; Reliability (semiconductor); Key (lock); Crash; Regularization (linguistics); Deep neural networks; Bayesian network; Collision; Data mining","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00005647202,0.0001632412,0.000197446,0.0001403701,0.0001153444,0.00007079211,0.00006227372,0.0001244333,0.00008785945],"category_scores_gemma":[0.000009671185,0.0001500683,0.00008243391,0.0004817185,0.00002552559,0.0001101799,0.0000317417,0.0002053717,4.964531e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005848971,"about_ca_system_score_gemma":0.000005382701,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001420885,"about_ca_topic_score_gemma":0.00004336786,"domain_scores_codex":[0.9992921,0.00002868697,0.0001865386,0.0001974557,0.0001151083,0.0001800589],"domain_scores_gemma":[0.9996746,0.00001021749,0.00003064811,0.0001848109,0.00002605629,0.0000737102],"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":[0.000006966163,0.000002990951,0.000209807,0.00001224652,0.0001353068,0.000006276993,0.00007515089,0.7576834,0.00004108464,0.0001495062,0.0004613645,0.2412159],"study_design_scores_gemma":[0.0001687074,0.00001267282,0.03396672,0.00001164419,0.0001692101,0.000008320611,0.00006289242,0.9645698,0.00002783515,0.00006488033,0.0007937307,0.0001435632],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07474671,0.0002743933,0.9217451,0.0000671481,0.000331526,0.00009815446,0.00001293863,0.00126565,0.001458435],"genre_scores_gemma":[0.9951235,0.0001150819,0.004328379,0.0001741135,0.00009896715,0.000002990232,0.0001255118,0.00001577498,0.00001567658],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9203768,"threshold_uncertainty_score":0.6119609,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.004332570029971204,"score_gpt":0.2145480504403708,"score_spread":0.2102154804103996,"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."}}