{"id":"W4390413835","doi":"10.1016/j.imavis.2023.104897","title":"Exploiting classifier inter-level features for efficient out-of-distribution detection","year":2023,"lang":"en","type":"article","venue":"Image and Vision Computing","topic":"Anomaly Detection Techniques and Applications","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Victoria; University of British Columbia, Okanagan Campus; University of British Columbia","funders":"Mitacs","keywords":"Computer science; Classifier (UML); Inference; Artificial intelligence; Exploit; Deep learning; Software deployment; Training set; Machine learning; Network architecture; Architecture; Pattern recognition (psychology); Data mining","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0003832409,0.0000889191,0.000111155,0.00009840856,0.0003247201,0.0001168977,0.0001721291,0.00005168575,7.783274e-7],"category_scores_gemma":[0.00006587685,0.00008371747,0.00006700886,0.0003564958,0.00003390789,0.0001244763,0.0002220379,0.00008720733,0.000006247261],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002083474,"about_ca_system_score_gemma":0.00001000554,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00000745952,"about_ca_topic_score_gemma":0.000001638685,"domain_scores_codex":[0.9991829,0.00002354176,0.0002306448,0.0002863793,0.0001034685,0.0001730999],"domain_scores_gemma":[0.9993401,0.0001647177,0.0001319705,0.0001878285,0.000134152,0.00004124196],"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":[0.000006216432,0.0000287847,0.00002122401,0.00003994257,0.000005745301,7.961103e-7,0.0003396618,0.0001082121,0.07295154,0.003607675,0.001402193,0.921488],"study_design_scores_gemma":[0.0002267092,0.0001433546,0.003843295,0.00007579032,0.000005423486,0.000007226881,0.0001765529,0.8542374,0.1365285,0.00121714,0.003394887,0.000143685],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06089613,0.00001554865,0.9379483,0.0002494918,0.000208852,0.0002021981,0.00001058015,0.0003811126,0.00008772605],"genre_scores_gemma":[0.9642998,0.000005863255,0.03547842,0.00003219901,0.00007966735,0.00001834121,0.00001159785,0.000007069111,0.00006702436],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9213443,"threshold_uncertainty_score":0.3413899,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03884751640672859,"score_gpt":0.3427141217430519,"score_spread":0.3038666053363234,"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."}}