{"id":"W4390414985","doi":"10.1016/j.neucom.2023.127183","title":"Analytically tractable heteroscedastic uncertainty quantification in Bayesian neural networks for regression tasks","year":2023,"lang":"en","type":"article","venue":"Neurocomputing","topic":"Gaussian Processes and Bayesian Inference","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":false,"ca_institutions":"Polytechnique Montréal","funders":"","keywords":"Homoscedasticity; Heteroscedasticity; Inference; Computer science; Covariate; Bayesian probability; Artificial neural network; Gaussian; Bayesian inference; Gaussian process; Artificial intelligence; Predictive inference; Regression; Mathematics; Uncertainty quantification; Machine learning; Statistics; Frequentist inference","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01032381,0.001554752,0.002496368,0.001282954,0.0009035147,0.003422193,0.003015212,0.003178833,0.003561292],"category_scores_gemma":[0.05077739,0.00185687,0.001310731,0.001704507,0.003006737,0.006338189,0.003586002,0.004794566,0.0004585801],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003713371,"about_ca_system_score_gemma":0.003727826,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01082953,"about_ca_topic_score_gemma":0.01251756,"domain_scores_codex":[0.996936,0.001652554,0.0001491241,0.0003919897,0.0006178847,0.0002524492],"domain_scores_gemma":[0.9720324,0.024819,0.0009580686,0.0006947354,0.00113685,0.0003588627],"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":[0.00006025315,0.00004887948,0.0004232804,0.0001644948,0.00005592848,0.00005909254,0.0001367631,0.7119742,0.0003416495,0.2673191,0.001602687,0.01781365],"study_design_scores_gemma":[0.00000468708,0.000005899972,0.00007139691,0.00002428111,0.00000726242,0.00001001382,0.000009796458,0.8488789,0.00008925389,0.1506644,0.0002249061,0.000009172036],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01027968,0.0008379237,0.9848503,0.0009638043,0.00003427322,0.00003356518,0.0001218684,0.00008736746,0.002791243],"genre_scores_gemma":[0.6951347,0.002790031,0.2896951,0.0006770715,0.0003862222,0.0004271538,0.0006164089,0.000358765,0.009914523],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01082953,"threshold_uncertainty_score":0.05459815,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03047839580974849,"score_gpt":0.2947843967770761,"score_spread":0.2643060009673276,"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."}}