{"id":"W3212356453","doi":"","title":"Fractal Structure and Generalization Properties of Stochastic Optimization Algorithms","year":2021,"lang":"en","type":"article","venue":"HAL (Le Centre pour la Communication Scientifique Directe)","topic":"Metaheuristic Optimization Algorithms Research","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Algorithm; Stochastic gradient descent; Ergodic theory; Generalization; Dynamical systems theory; Mathematics; Computer science; Hessian matrix; Fractal; Artificial neural network; Bounded function; Hyperparameter; Invariant measure; Mathematical optimization; Artificial intelligence; Applied mathematics","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":[0.001384929,0.0004104655,0.0008369357,0.001889006,0.00071414,0.001552042,0.0004845097,0.0007323477,0.002616626],"category_scores_gemma":[0.01170945,0.0004081287,0.0008719338,0.001095433,0.001419334,0.001697152,0.0008302653,0.001193587,0.0002094724],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001004225,"about_ca_system_score_gemma":0.0005897667,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001421044,"about_ca_topic_score_gemma":0.0008793343,"domain_scores_codex":[0.9994647,0.0001766524,0.00003830672,0.00008219614,0.0001822033,0.00005596903],"domain_scores_gemma":[0.9924387,0.005237979,0.0007744292,0.0005068916,0.0007827275,0.0002593031],"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":[0.00004296845,0.00003136996,0.001704496,0.00009426721,0.00004204087,0.0001124332,0.0001593935,0.1795644,0.00211254,0.7944641,0.001593903,0.02007811],"study_design_scores_gemma":[0.00001554682,0.00003647352,0.001552584,0.00002760109,0.00001786967,0.0001338305,0.0000248035,0.7048751,0.0004620928,0.2908736,0.001964643,0.0000158406],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3409623,0.003253121,0.6138521,0.001565417,0.0002313794,0.00009256513,0.0002354502,0.0004242217,0.03938343],"genre_scores_gemma":[0.9369273,0.001329475,0.05517244,0.0001624955,0.0003246324,0.0001011125,0.0002286596,0.0001377618,0.005616031],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002616626,"threshold_uncertainty_score":0.008753419,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0154306595846094,"score_gpt":0.2283326279716174,"score_spread":0.212901968387008,"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."}}