{"id":"W2920348561","doi":"10.48550/arxiv.1903.02639","title":"IMEXnet: A Forward Stable Deep Neural Network","year":2019,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Model Reduction and Neural Networks","field":"Physics and Astronomy","cited_by":24,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Computer science; Robustness (evolution); Residual; Convolutional neural network; Artificial intelligence; Convolution (computer science); Generalization; Key (lock); Segmentation; Artificial neural network; Deep learning; Sensitivity (control systems); Pixel; Limit (mathematics); Field (mathematics); Algorithm; Machine learning; Pattern recognition (psychology); 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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0001014486,0.0003911386,0.0004361677,0.00008214014,0.00017048,0.00008479997,0.0005322359,0.0002054152,0.001412269],"category_scores_gemma":[0.000001329447,0.0004372977,0.0004003372,0.0002877891,0.00006639519,0.0001780168,0.0007009343,0.0008708493,0.0002613453],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007128373,"about_ca_system_score_gemma":0.00007954854,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002513431,"about_ca_topic_score_gemma":0.00001237846,"domain_scores_codex":[0.9981632,0.0001018673,0.0002102572,0.0008706815,0.0000762694,0.0005777019],"domain_scores_gemma":[0.9986333,0.00005041635,0.0002528393,0.0007706158,0.00008787786,0.0002049365],"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":[0.00005858633,0.00004895647,0.01174687,0.00002185589,0.0001418472,0.00001273837,0.00003160781,0.953773,0.000003874129,0.02907102,0.003900163,0.001189502],"study_design_scores_gemma":[0.0005163488,0.00003284328,0.0003871794,0.00004310021,0.0001433833,9.223868e-7,0.00009909145,0.9644601,0.0000223142,0.02830797,0.005454191,0.0005325642],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6552024,0.0001819309,0.299567,0.00009985589,0.003003331,0.0007836648,0.00005936675,0.0001986195,0.04090388],"genre_scores_gemma":[0.9869192,0.0000357325,0.0001046936,0.00008701581,0.001093969,0.000002030576,0.0001314698,0.00004420405,0.01158174],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3317167,"threshold_uncertainty_score":0.9998079,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04425069276499816,"score_gpt":0.1820515040665877,"score_spread":0.1378008113015896,"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."}}