{"id":"W3015482436","doi":"10.1016/j.neunet.2020.04.001","title":"A tight upper bound on the generalization error of feedforward neural networks","year":2020,"lang":"en","type":"article","venue":"Neural Networks","topic":"Machine Learning and ELM","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":false,"ca_institutions":"Ericsson (Canada)","funders":"","keywords":"Differentiable function; Generalization; Upper and lower bounds; Term (time); Artificial neural network; Feed forward; Feedforward neural network; Mathematics; Sensitivity (control systems); Computer science; Function (biology); Generalization error; Space (punctuation); Algorithm; Applied mathematics; Artificial intelligence; Mathematical analysis; Physics","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":[0.0001906534,0.0002100447,0.0002241152,0.00003425668,0.0002428088,0.0001780848,0.0008900671,0.0001060804,0.00005502305],"category_scores_gemma":[0.00005227998,0.0001397188,0.0001390836,0.000593644,0.00006371612,0.0001878998,0.0002316742,0.0004382978,0.00001172915],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001276207,"about_ca_system_score_gemma":0.00001140208,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003060966,"about_ca_topic_score_gemma":0.000006356875,"domain_scores_codex":[0.9983642,0.0002539281,0.0002938813,0.0003983752,0.0002976459,0.0003920286],"domain_scores_gemma":[0.9990124,0.000163826,0.0001684551,0.0004604121,0.00005591581,0.0001389512],"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":[0.00002767875,0.00001861959,0.001874853,0.000005430662,0.00001372573,0.000006325152,0.0002699839,0.9553203,0.00001684778,0.006823319,0.01421701,0.0214059],"study_design_scores_gemma":[0.0002046188,0.0002047214,0.003092445,0.00001081376,0.000009264672,0.00000614077,0.000005793902,0.9918013,0.00002038889,0.00007812937,0.00441513,0.0001512973],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2963875,0.0006566839,0.6465827,0.05161785,0.001843432,0.0004539245,0.000001899623,0.0004524667,0.002003561],"genre_scores_gemma":[0.9871936,0.00001694172,0.0003704605,0.01138631,0.0008289238,0.000009436749,0.000010318,0.00002062288,0.0001633677],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6908061,"threshold_uncertainty_score":0.5697569,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02096427600500496,"score_gpt":0.2424646085636059,"score_spread":0.2215003325586009,"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."}}