{"id":"W4403016954","doi":"10.1016/j.ijar.2024.109301","title":"Uncertainty-based knowledge distillation for Bayesian deep neural network compression","year":2024,"lang":"en","type":"article","venue":"International Journal of Approximate Reasoning","topic":"Anomaly Detection Techniques and Applications","field":"Computer Science","cited_by":10,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Windsor","funders":"","keywords":"Artificial neural network; Computer science; Bayesian network; Artificial intelligence; Bayesian probability; Distillation; Machine learning; Compression (physics); Chemistry; Chromatography; Materials science","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":[0.0004122263,0.0001119534,0.000137274,0.0001761192,0.0001189484,0.000379034,0.0006280483,0.00005305178,0.00001019973],"category_scores_gemma":[0.00004215015,0.00009339276,0.0001799419,0.0002379743,0.0000254849,0.0003998454,0.00007401277,0.0001622183,0.000002660149],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001305823,"about_ca_system_score_gemma":0.00007841751,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000003520562,"about_ca_topic_score_gemma":0.000001823235,"domain_scores_codex":[0.9989716,0.00003505117,0.000388747,0.0001886666,0.0002579003,0.0001579826],"domain_scores_gemma":[0.9989306,0.0002237219,0.0002362389,0.0001314005,0.000405513,0.00007247938],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0000956267,0.0001057476,0.0002831031,0.00006620984,0.0001371415,0.0000303081,0.0003443581,0.1776338,0.001142698,0.2881118,0.002233173,0.529816],"study_design_scores_gemma":[0.0001937233,0.00006848226,0.0001243681,0.0002544177,0.00001177134,0.00009439522,0.00001107336,0.968015,0.0008428522,0.009895317,0.02039183,0.0000968076],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.002376269,0.0005887508,0.9940157,0.001095615,0.001022166,0.0001428478,0.000004652451,0.0001552468,0.0005987331],"genre_scores_gemma":[0.8050662,0.0000168444,0.1941612,0.0000550992,0.0006266303,0.00002195125,0.00000900427,0.00001227978,0.00003070283],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.80269,"threshold_uncertainty_score":0.3808446,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0136425166499591,"score_gpt":0.3046256763103273,"score_spread":0.2909831596603683,"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."}}