{"id":"W4319339632","doi":"10.36227/techrxiv.22006742","title":"Improving Image Recognition Accuracy in Neural Networks Using Fractional Natural Gradient Descent","year":2023,"lang":"en","type":"preprint","venue":"","topic":"Image Processing Techniques and Applications","field":"Engineering","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Ministry of Science and Higher Education of the Russian Federation; Russian Science Foundation; Centre de Recherches Mathématiques","keywords":"MNIST database; Convexity; Gradient descent; Mathematics; Rate of convergence; Curvature; Convergence (economics); Momentum (technical analysis); Artificial neural network; Image (mathematics); Order (exchange); Applied mathematics; Algorithm; Computer science; Artificial intelligence; Geometry","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001040822,0.0006336059,0.0007486655,0.0006249337,0.0003629373,0.0009271805,0.0008085785,0.0008955955,0.001292499],"category_scores_gemma":[0.003446831,0.0002601721,0.000444924,0.0005612265,0.0005758848,0.001354907,0.0005349643,0.0006883522,0.0004632443],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007712723,"about_ca_system_score_gemma":0.0005117852,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004730639,"about_ca_topic_score_gemma":0.00369717,"domain_scores_codex":[0.9995776,0.0001085822,0.00002895584,0.0001082439,0.0001260763,0.00005052088],"domain_scores_gemma":[0.9993423,0.0003109498,0.00006726833,0.0001118201,0.0001432965,0.00002432611],"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":[0.0003144136,0.0001209846,0.001469907,0.0001166279,0.0001034959,0.0002006567,0.0001137578,0.4906395,0.0350078,0.009189511,0.003668896,0.4590544],"study_design_scores_gemma":[0.000003993056,0.00001990319,0.0002143629,0.000002817083,0.000005251262,0.0000268854,0.00000280195,0.9937743,0.004132834,0.001411332,0.0004010563,0.000004478896],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06901243,0.0007431914,0.9258571,0.0002802798,0.0001385871,0.00003217908,0.00004544968,0.001510548,0.002380144],"genre_scores_gemma":[0.64991,0.0003799308,0.3451971,0.000211248,0.0001015236,0.00004561624,0.0001270546,0.0002263676,0.003801098],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004730639,"threshold_uncertainty_score":0.009406209,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04066706910852975,"score_gpt":0.2923875884603741,"score_spread":0.2517205193518444,"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."}}