{"id":"W2949190276","doi":"","title":"On the difficulty of training Recurrent Neural Networks","year":2012,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Neural Networks and Applications","field":"Computer Science","cited_by":140,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Constraint (computer-aided design); Perspective (graphical); Computer science; Artificial neural network; Simple (philosophy); Norm (philosophy); Artificial intelligence; Gradient descent; Mathematical optimization; Algorithm; Applied mathematics; Mathematics; Geometry; Epistemology","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":[0.0001687753,0.0002081376,0.0002156892,0.00005189482,0.0001606939,0.00005029282,0.001650132,0.0001311556,0.00001673967],"category_scores_gemma":[0.00001088182,0.000160111,0.000194102,0.0004047268,0.00009630374,0.0001071848,0.001005951,0.0006418378,0.000009700831],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004233395,"about_ca_system_score_gemma":0.00002755613,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001489359,"about_ca_topic_score_gemma":0.00000639882,"domain_scores_codex":[0.9988281,0.0001145951,0.0001646229,0.0005165026,0.00007038518,0.000305811],"domain_scores_gemma":[0.9982595,0.0002908367,0.0002617836,0.001020018,0.00006118655,0.0001066043],"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":[0.000005279468,0.00005951638,0.0001045659,0.000006554753,0.00002037978,0.000004975371,0.0001492858,0.5043672,0.000005788212,0.4905089,0.00065002,0.004117513],"study_design_scores_gemma":[0.00009714736,0.00002420111,0.00110259,0.00004858106,0.00002651872,0.000001393128,0.00002937135,0.9875546,0.00001267714,0.01069216,0.0002210639,0.000189765],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4255062,0.00008327095,0.5712043,0.0005502145,0.0008123084,0.0003665476,0.000009165301,0.0001196904,0.001348272],"genre_scores_gemma":[0.9993502,0.00006802109,0.0001348274,0.0001382865,0.0001448371,0.000002249695,0.000007808053,0.000008889551,0.0001448698],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.573844,"threshold_uncertainty_score":0.6529137,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1297134932191586,"score_gpt":0.2002915810523392,"score_spread":0.07057808783318056,"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."}}