{"id":"W3092726106","doi":"10.48550/arxiv.2010.08097","title":"Consistent Feature Selection for Analytic Deep Neural Networks","year":2020,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Statistical Methods and Inference","field":"Mathematics","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University","funders":"","keywords":"Interpretability; Artificial intelligence; Artificial neural network; Computer science; Deep learning; Machine learning; Feature selection; Feature (linguistics); Convolutional neural network; Selection (genetic algorithm); Lasso (programming language); Consistency (knowledge bases); Estimator; Class (philosophy); Mathematics","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":[0.01009653,0.001013039,0.001431187,0.001358825,0.0006194772,0.001249764,0.001545136,0.001259965,0.001695467],"category_scores_gemma":[0.02926268,0.0006762394,0.001011249,0.001019359,0.002325292,0.001736843,0.002067736,0.002600021,0.0003794764],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001189397,"about_ca_system_score_gemma":0.001421913,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00111064,"about_ca_topic_score_gemma":0.001006873,"domain_scores_codex":[0.9952902,0.003057614,0.0001591321,0.0005162728,0.0007786909,0.0001981421],"domain_scores_gemma":[0.9862299,0.01034195,0.0009672705,0.001041603,0.001208831,0.0002105643],"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":[0.0003044518,0.0001108378,0.004477698,0.0002715251,0.0002533694,0.000334999,0.0002507755,0.5568702,0.003816134,0.3245749,0.0052107,0.1035243],"study_design_scores_gemma":[0.00002324616,0.00003361094,0.0002433661,0.00002045116,0.000009965684,0.00002455291,0.00001649038,0.8452047,0.0006278287,0.1531202,0.0006674023,0.000008031927],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.009277021,0.000221959,0.989306,0.0004689767,0.00001952947,0.00003485704,0.00006928224,0.0001564692,0.0004458762],"genre_scores_gemma":[0.6190975,0.0006764112,0.3742198,0.0006375583,0.0002204718,0.000748864,0.000835035,0.000287501,0.003276856],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01009653,"threshold_uncertainty_score":0.05339622,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2135978634310282,"score_gpt":0.2729604427719731,"score_spread":0.05936257934094491,"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."}}