{"id":"W4321012054","doi":"10.48550/arxiv.2302.06763","title":"Breaking the Lower Bound with (Little) Structure: Acceleration in Non-Convex Stochastic Optimization with Heavy-Tailed Noise","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Stochastic Gradient Optimization Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"York University; National Science Foundation","keywords":"Bounded function; Mathematics; Upper and lower bounds; Combinatorics; Omega; Convex function; Rate of convergence; Regular polygon; Discrete mathematics; Physics; Mathematical analysis; Quantum mechanics; Computer science; 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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0002485841,0.0004515112,0.0003808903,0.0005545056,0.0002701071,0.0004388107,0.001533527,0.0002642989,0.00002444184],"category_scores_gemma":[0.00004416189,0.0003639687,0.00007191033,0.001777349,0.0002006075,0.0006670294,0.0008115019,0.0006535889,0.000009097585],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003827091,"about_ca_system_score_gemma":0.0003324922,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002177732,"about_ca_topic_score_gemma":0.0003439998,"domain_scores_codex":[0.9976711,0.0001188604,0.0002945877,0.001239489,0.0002491199,0.0004268839],"domain_scores_gemma":[0.9976307,0.0001827221,0.0004284684,0.00130864,0.0003416967,0.0001078133],"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":[0.0001183874,0.0000551552,0.0006343538,0.00002900355,0.00005025861,0.00008575007,0.0004543849,0.9864419,0.000006095565,0.01202644,0.00003280808,0.00006545444],"study_design_scores_gemma":[0.0007172506,0.0001802384,0.001327348,0.0002894486,0.00006276234,0.000011811,0.00007000505,0.9912014,0.00005868112,0.00561061,0.000001947945,0.0004685072],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04520654,0.000008057385,0.9524187,0.0002055484,0.0003664921,0.001185563,0.00001233494,0.0004806086,0.0001161464],"genre_scores_gemma":[0.9800276,0.00001140032,0.019494,0.0000895648,0.00005118187,0.00001377809,0.00005577605,0.00005300072,0.0002036859],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9348211,"threshold_uncertainty_score":0.9998812,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03945059861527851,"score_gpt":0.1882352903061463,"score_spread":0.1487846916908678,"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."}}