{"id":"W2965944281","doi":"","title":"Tight analyses for non-smooth stochastic gradient descent","year":2019,"lang":"en","type":"article","venue":"Conference on Learning Theory","topic":"Stochastic Gradient Optimization Techniques","field":"Computer Science","cited_by":41,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Mathematics; Lipschitz continuity; Stochastic gradient descent; Differentiable function; Combinatorics; Convex function; Gradient descent; Upper and lower bounds; Regular polygon; Discrete mathematics; Applied mathematics; Mathematical analysis; Computer science","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":[0.02689398,0.005478594,0.005683759,0.004258015,0.003440413,0.005118314,0.006526676,0.00547856,0.01155368],"category_scores_gemma":[0.1309287,0.002581611,0.005286602,0.002817784,0.007728141,0.01225153,0.01086107,0.0144837,0.002261876],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.007805124,"about_ca_system_score_gemma":0.005633266,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008286189,"about_ca_topic_score_gemma":0.006163376,"domain_scores_codex":[0.9879396,0.004328775,0.0006049093,0.002113406,0.003738483,0.001274775],"domain_scores_gemma":[0.9178297,0.05936402,0.005036654,0.007683893,0.007581455,0.002504308],"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":[0.0004304193,0.000189938,0.003074142,0.0008487701,0.0003434249,0.0003481624,0.0005948542,0.3736889,0.003638049,0.5788928,0.009510169,0.02844047],"study_design_scores_gemma":[0.00002887607,0.0001121183,0.0006552244,0.0001649774,0.00007747229,0.00007474286,0.00004978207,0.7913679,0.001172443,0.2033768,0.002871656,0.00004804743],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01173335,0.003163206,0.9712432,0.002678832,0.0002855611,0.0001271436,0.0002373122,0.0006427363,0.009888599],"genre_scores_gemma":[0.5875375,0.006520189,0.3614404,0.00516221,0.001881202,0.001473719,0.001625798,0.003211356,0.0311477],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.02689398,"threshold_uncertainty_score":0.1422306,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04690993015189428,"score_gpt":0.3102476221312967,"score_spread":0.2633376919794024,"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."}}