{"id":"W2587753047","doi":"10.48550/arxiv.1702.00071","title":"On orthogonality and learning recurrent networks with long term dependencies","year":2017,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Sparse and Compressive Sensing Techniques","field":"Engineering","cited_by":116,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal","funders":"","keywords":"Orthogonality; Backpropagation; Artificial neural network; Term (time); Convergence (economics); Stability (learning theory); Computer science; Matrix (chemical analysis); Property (philosophy); Norm (philosophy); Gradient descent; Applied mathematics; Mathematical optimization; Mathematics; Artificial intelligence; Machine learning","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":[0.0001163371,0.0003000892,0.0002826987,0.0001016197,0.0002325204,0.0001106394,0.0003190066,0.0002376285,0.00001017434],"category_scores_gemma":[0.00001829718,0.0003080087,0.00007119086,0.00005485435,0.0001181145,0.00009407421,0.0003625294,0.001158967,0.000003818069],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008356601,"about_ca_system_score_gemma":0.00002362925,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004729837,"about_ca_topic_score_gemma":0.0001336776,"domain_scores_codex":[0.9990378,0.0000663488,0.00009609457,0.0004921861,0.00006288727,0.0002447025],"domain_scores_gemma":[0.9990706,0.00007260406,0.0001144544,0.0005851327,0.00006217779,0.0000950452],"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":[0.00007644472,0.00001929904,0.03053208,0.00007083381,0.0001506801,0.0004587645,0.00005723179,0.9627531,0.00001094536,0.002017362,0.0001144428,0.003738854],"study_design_scores_gemma":[0.0005881569,0.0002312556,0.1151188,0.00209588,0.0003041771,0.00002996239,0.00004820837,0.8742313,0.0003644362,0.005453941,0.0002393179,0.001294622],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9060337,0.0003205526,0.08909679,0.000004808386,0.0002040132,0.0001635965,0.000003551307,0.000607672,0.0035653],"genre_scores_gemma":[0.9985492,0.0009109286,0.0001394674,0.000008758994,0.00006601871,7.355315e-7,0.00001807174,0.00003517147,0.000271669],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.09251546,"threshold_uncertainty_score":0.9999372,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05629598084733464,"score_gpt":0.1876238132149216,"score_spread":0.131327832367587,"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."}}