{"id":"W2964309400","doi":"","title":"Revisiting Natural Gradient for Deep Networks","year":2014,"lang":"en","type":"article","venue":"","topic":"Neural Networks and Applications","field":"Computer Science","cited_by":139,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Hessian matrix; Stochastic gradient descent; Gradient descent; Robustness (evolution); Diagonal; Generalization; Computer science; Metric (unit); Deep learning; Subspace topology; Algorithm; Artificial intelligence; Mathematics; Mathematical optimization; Applied mathematics; Artificial neural network","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":[0.006318151,0.001675417,0.001171157,0.001464978,0.0008227623,0.001937466,0.002234035,0.001932133,0.00371964],"category_scores_gemma":[0.02533026,0.0005898904,0.0007791548,0.0009097101,0.001902106,0.005234388,0.002895928,0.002675071,0.000792396],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002063322,"about_ca_system_score_gemma":0.002714969,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007159683,"about_ca_topic_score_gemma":0.01106919,"domain_scores_codex":[0.9976845,0.00103306,0.0001143641,0.0003604569,0.0006834241,0.0001242048],"domain_scores_gemma":[0.9932489,0.003721466,0.0003506915,0.0009866835,0.001435481,0.000256696],"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":[0.000349217,0.0002613606,0.003299397,0.0003587898,0.0001518002,0.0001023948,0.0001525284,0.7114522,0.003473282,0.09990432,0.008876719,0.171618],"study_design_scores_gemma":[0.00002282798,0.00008104837,0.000154296,0.00002405144,0.000007978289,0.00002543305,0.00001123241,0.9716665,0.001296335,0.02508413,0.001616548,0.000009541352],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.03725885,0.001542451,0.9501421,0.0008478009,0.0002935837,0.0001957553,0.000338942,0.002521598,0.006858971],"genre_scores_gemma":[0.4179019,0.0006015387,0.575613,0.0005444742,0.0001605542,0.0002411266,0.0006761843,0.0007861305,0.003475022],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.007159683,"threshold_uncertainty_score":0.03341401,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.009799498980287287,"score_gpt":0.2366033906371499,"score_spread":0.2268038916568626,"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."}}