{"id":"W1844261860","doi":"10.48550/arxiv.1301.3584","title":"Revisiting Natural Gradient for Deep Networks","year":2013,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Stochastic Gradient Optimization Techniques","field":"Computer Science","cited_by":122,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"Natural Sciences and Engineering Research Council of Canada; DeepMind; Compute Canada","keywords":"Natural (archaeology); Computer science; Artificial intelligence; Geology; Paleontology","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0002488052,0.0003465981,0.0003519737,0.0002758539,0.0002116942,0.0002280116,0.001957289,0.000267423,0.00001321476],"category_scores_gemma":[0.00009892904,0.0003957014,0.0002755234,0.0004758696,0.00007775766,0.0003771284,0.001685899,0.0004950552,0.00001607843],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000252219,"about_ca_system_score_gemma":0.00006220053,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003325224,"about_ca_topic_score_gemma":0.000002876273,"domain_scores_codex":[0.9979527,0.0000740993,0.0002665358,0.001144244,0.00007977324,0.0004826764],"domain_scores_gemma":[0.9977466,0.0002226951,0.0003816529,0.001118056,0.0003707552,0.0001602117],"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":[0.00000624485,0.00001864764,0.0001294506,0.00005013058,0.00004381841,0.00001627268,0.00008033736,0.4973638,0.000002297448,0.4992869,0.000436723,0.0025654],"study_design_scores_gemma":[0.0002262465,0.00003177433,0.0001042765,0.0001201419,0.00003961826,0.00000406651,0.00001525059,0.942128,0.00002480735,0.05676876,0.00013593,0.0004011253],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.004412605,0.0002189185,0.991865,0.0001093403,0.001193246,0.0009543546,0.000004002484,0.0007332132,0.0005092848],"genre_scores_gemma":[0.9029092,0.00008853307,0.09605441,0.0001458479,0.0001881617,0.00001306756,0.00003631434,0.00002676628,0.0005376695],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8984966,"threshold_uncertainty_score":0.9998495,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04715239288676829,"score_gpt":0.1881851438096175,"score_spread":0.1410327509228492,"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."}}