{"id":"W2178103884","doi":"10.48550/arxiv.1511.08400","title":"Regularizing RNNs by Stabilizing Activations","year":2015,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":30,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"Air Force Research Laboratory; Advanced Research Projects Agency; Defense Advanced Research Projects Agency; U.S. Department of Defense","keywords":"Recurrent neural network; TIMIT; Computer science; Term (time); Dropout (neural networks); Speech recognition; Task (project management); Language model; Artificial intelligence; Artificial neural network; Machine learning; Hidden Markov model","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0003473599,0.0002605886,0.0002782661,0.000259141,0.0001905077,0.0002126087,0.001423101,0.0002761733,0.00007643049],"category_scores_gemma":[0.0001340391,0.0003136393,0.0001735288,0.0005630199,0.00006258526,0.0005558459,0.001296081,0.0004794039,0.0001470869],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003390917,"about_ca_system_score_gemma":0.0002782139,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001190793,"about_ca_topic_score_gemma":0.0000216087,"domain_scores_codex":[0.9982509,0.000170706,0.0001842913,0.0009458914,0.0001306351,0.0003175737],"domain_scores_gemma":[0.9980081,0.0001545112,0.0002163084,0.001106654,0.0002769904,0.0002373964],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001421623,0.001891208,0.006790345,0.0004910979,0.001358926,0.001270325,0.00605848,0.04090518,0.009819313,0.6833051,0.1685584,0.07940942],"study_design_scores_gemma":[0.001211372,0.00008656951,0.0004887052,0.0003302274,0.0001801453,0.0000206018,0.001412245,0.6791205,0.009955195,0.2794183,0.02570431,0.002071849],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1438855,0.00005683826,0.8395686,0.0005263904,0.0005812444,0.0002612713,0.00004071874,0.0005270877,0.01455228],"genre_scores_gemma":[0.9867766,0.0000569013,0.01068497,0.000142196,0.00005736118,0.000001579979,0.00003230121,0.00001964198,0.002228479],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.842891,"threshold_uncertainty_score":0.9999316,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1401149921742679,"score_gpt":0.1975755539048891,"score_spread":0.05746056173062122,"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."}}