{"id":"W1533861849","doi":"","title":"Understanding the difficulty of training deep feedforward neural networks","year":2010,"lang":"en","type":"article","venue":"","topic":"Neural Networks and Applications","field":"Computer Science","cited_by":12692,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Initialization; Computer science; Artificial neural network; Artificial intelligence; Deep neural networks; Deep learning; Gradient descent; Jacobian matrix and determinant; Sigmoid function; Machine learning; 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":[0.004251104,0.001231306,0.0008083709,0.0005593206,0.0005878959,0.001697693,0.001257948,0.002456123,0.002333204],"category_scores_gemma":[0.03511711,0.001132296,0.0004449435,0.0006012978,0.002168092,0.00735236,0.001444381,0.003728366,0.0006827879],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001152828,"about_ca_system_score_gemma":0.0009506122,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004214759,"about_ca_topic_score_gemma":0.003014095,"domain_scores_codex":[0.9987323,0.0005057075,0.0001198882,0.0002780401,0.0002768587,0.00008723133],"domain_scores_gemma":[0.9844468,0.01308933,0.0006046043,0.0006283273,0.001056707,0.0001742129],"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":[0.0003069462,0.00008840147,0.005512543,0.0009657211,0.0001128905,0.0003217843,0.0005437847,0.7178414,0.008650141,0.06050633,0.005986174,0.1991639],"study_design_scores_gemma":[0.00002961156,0.00007435866,0.001389644,0.0001256148,0.00001885873,0.0001412341,0.00009834336,0.8830338,0.004467994,0.10736,0.003230851,0.00002977859],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07508201,0.005425712,0.9025662,0.008326568,0.0002646802,0.00009511194,0.0002358146,0.0009108178,0.007093107],"genre_scores_gemma":[0.7114234,0.005950375,0.2748507,0.001075446,0.0006359874,0.0002507753,0.0005459769,0.0004405935,0.004826802],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004251104,"threshold_uncertainty_score":0.02248228,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0766112791778204,"score_gpt":0.2560151960910123,"score_spread":0.1794039169131919,"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."}}