{"id":"W2995434136","doi":"10.48550/arxiv.1905.12558","title":"Limitations of the Empirical Fisher Approximation for Natural Gradient\\n Descent","year":2019,"lang":"","type":"preprint","venue":"arXiv (Cornell University)","topic":"Stochastic Gradient Optimization Techniques","field":"Computer Science","cited_by":30,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Fisher information; Hessian matrix; Fisher kernel; Heuristics; Mathematics; Applied mathematics; Econometrics; Mathematical economics; Mathematical optimization; Statistics; Computer science; Artificial intelligence; Kernel method","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":[0.008036,0.0009127085,0.001545374,0.000774933,0.0009248679,0.002202916,0.002986463,0.001911738,0.005089191],"category_scores_gemma":[0.05639027,0.0009035785,0.0006350057,0.0008733316,0.00281831,0.004820098,0.002853913,0.004406353,0.002245564],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001711759,"about_ca_system_score_gemma":0.003147207,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008521223,"about_ca_topic_score_gemma":0.008344332,"domain_scores_codex":[0.9956701,0.002187654,0.0001797032,0.0004486101,0.001324742,0.0001891736],"domain_scores_gemma":[0.9824599,0.01361001,0.0003314855,0.002106299,0.001245496,0.0002467728],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0002575175,0.00007128329,0.001224604,0.0003802511,0.00007632338,0.0001149431,0.0002516664,0.2055474,0.0007073301,0.6694352,0.01903802,0.1028955],"study_design_scores_gemma":[0.00002369764,0.00002366534,0.0001766798,0.00008041195,0.000005132976,0.00006678564,0.00002876112,0.7947703,0.0004127332,0.1973376,0.007058755,0.00001547462],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.007309853,0.002094635,0.9705346,0.004448213,0.0002742003,0.00005649168,0.0001589834,0.0008505746,0.0142724],"genre_scores_gemma":[0.4152451,0.004171542,0.5582514,0.002158464,0.0006252193,0.000484791,0.0006858023,0.001140672,0.01723701],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.008521223,"threshold_uncertainty_score":0.04249895,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2954199039827467,"score_gpt":0.2289787568300307,"score_spread":0.06644114715271598,"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."}}