{"id":"W2963603213","doi":"","title":"Efficient Exact Gradient Update for training Deep Networks with Very Large Sparse Targets.","year":2015,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":28,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal; Computer Research Institute of Montréal; Canadian Institute for Advanced Research","funders":"","keywords":"Softmax function; Backpropagation; Computer science; Artificial neural network; Speedup; Dimension (graph theory); Algorithm; Deep learning; Computation; Sparse matrix; Gradient descent; Artificial intelligence; Mathematics; Parallel computing","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":[0.0006556185,0.0001609959,0.0001880607,0.0000665837,0.0001065488,0.0001140837,0.000528024,0.00005050006,0.000008845044],"category_scores_gemma":[0.0000255196,0.0001214904,0.00005515391,0.0001890153,0.00001939576,0.0001365504,0.000164332,0.0001057831,0.00001472875],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007045,"about_ca_system_score_gemma":0.00008604403,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001224326,"about_ca_topic_score_gemma":0.00004767317,"domain_scores_codex":[0.9984472,0.00003099221,0.0002092943,0.0004585235,0.0002604405,0.0005935265],"domain_scores_gemma":[0.9990252,0.00005206214,0.00006896182,0.0005066273,0.00009874815,0.0002483489],"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":[0.00002258776,0.00008064327,0.0001855836,0.000005218258,0.00001975416,0.00001520913,0.002133299,0.9191078,0.000001994518,0.06653319,0.0005786258,0.0113161],"study_design_scores_gemma":[0.0008703675,0.0001271591,0.0000395365,0.00001168006,0.00000681874,0.00001272925,0.0001945613,0.9893867,0.00002600593,0.0006542309,0.008478409,0.0001917679],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0262558,0.0001807038,0.9707464,0.0003716596,0.0005822916,0.0002795628,0.000001068693,0.0002114764,0.001371043],"genre_scores_gemma":[0.7450822,0.000001545322,0.2541409,0.0005151971,0.0001138427,0.00002889913,0.000003910909,0.00001211762,0.0001014326],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7188264,"threshold_uncertainty_score":0.4954233,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04118044383817406,"score_gpt":0.2430355192364063,"score_spread":0.2018550753982322,"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."}}