{"id":"W2981902149","doi":"10.48550/arxiv.1910.10307","title":"Detecting Out-of-Distribution Inputs in Deep Neural Networks Using an Early-Layer Output","year":2019,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Anomaly Detection Techniques and Applications","field":"Computer Science","cited_by":35,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Layer (electronics); Artificial neural network; Distribution (mathematics); Computer science; Deep neural networks; Artificial intelligence; Mathematics; Materials science; Nanotechnology","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":[0.002474752,0.00181795,0.001389161,0.001801044,0.0005968687,0.001480309,0.001858437,0.001802462,0.001070033],"category_scores_gemma":[0.007369752,0.0004745767,0.0007092887,0.001217947,0.001018081,0.002517341,0.001972342,0.002893827,0.0008254325],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001324699,"about_ca_system_score_gemma":0.0008156558,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003344791,"about_ca_topic_score_gemma":0.005381948,"domain_scores_codex":[0.9982958,0.0002753968,0.0001028038,0.0004677336,0.0006116882,0.0002465434],"domain_scores_gemma":[0.9957956,0.001890119,0.0005663138,0.0006071206,0.0009666489,0.0001741164],"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":[0.0009450197,0.0006590154,0.04041666,0.0002779003,0.0002519893,0.0004096394,0.000211466,0.137565,0.04175455,0.004048988,0.007223332,0.7662366],"study_design_scores_gemma":[0.00001357205,0.00009653259,0.003820274,0.00002666788,0.00002874655,0.0001386479,0.0000428871,0.967064,0.02347546,0.004179838,0.001094292,0.00001905744],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2148143,0.001551354,0.7743457,0.0006689947,0.0002278776,0.0001030876,0.0004943564,0.005379631,0.002414693],"genre_scores_gemma":[0.8245441,0.0004831836,0.169738,0.0003119138,0.00009858049,0.00008173217,0.001026835,0.0002686484,0.003446886],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003344791,"threshold_uncertainty_score":0.01308793,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09811232229998548,"score_gpt":0.2244071822298095,"score_spread":0.126294859929824,"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."}}