{"id":"W4321636690","doi":"10.1109/sc41404.2022.00052","title":"HyLo: A Hybrid Low-Rank Natural Gradient Descent Method","year":2022,"lang":"en","type":"article","venue":"","topic":"Sparse and Compressive Sensing Techniques","field":"Engineering","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Kronecker delta; Scalability; Gradient descent; Computer science; Quadratic growth; Computation; Factorization; Algorithm; Stochastic gradient descent; Matrix decomposition; Residual neural network; Rank (graph theory); Artificial neural network; Mathematics; Artificial intelligence; Combinatorics","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":[0.000806452,0.0009331866,0.001116266,0.0005688635,0.0004734809,0.0009893615,0.001926447,0.001271831,0.007407332],"category_scores_gemma":[0.002447502,0.0004941283,0.0005839537,0.0004838652,0.0005869343,0.001298724,0.001292794,0.001428907,0.00289122],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007616582,"about_ca_system_score_gemma":0.002463353,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01095398,"about_ca_topic_score_gemma":0.02219677,"domain_scores_codex":[0.999553,0.0001100805,0.00002387287,0.00008979985,0.0001509235,0.0000722391],"domain_scores_gemma":[0.9993508,0.0002292214,0.00004964821,0.0001179004,0.0001869982,0.00006536356],"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":[0.0005787722,0.0003731405,0.001506772,0.0003425282,0.0001878179,0.0002098237,0.0001497187,0.4732076,0.01256735,0.02361933,0.04468171,0.4425755],"study_design_scores_gemma":[0.00004753848,0.0000344069,0.00008061803,0.000006224895,0.000004674715,0.00001696597,0.000007876572,0.994387,0.001006045,0.002137803,0.002262603,0.000008248195],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01022231,0.000338984,0.9805439,0.0002573116,0.0001053671,0.0001042337,0.0002728118,0.00535467,0.002800373],"genre_scores_gemma":[0.1444286,0.0002095555,0.8400983,0.0006638302,0.0000976777,0.0004035441,0.00157492,0.001199381,0.01132416],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01095398,"threshold_uncertainty_score":0.02478004,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.009293411424507457,"score_gpt":0.2278365780491658,"score_spread":0.2185431666246584,"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."}}