{"id":"W2947651699","doi":"10.48550/arxiv.1905.13200","title":"Exploiting Uncertainty of Loss Landscape for Stochastic Optimization","year":2019,"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":"University of Toronto","funders":"","keywords":"Stochastic optimization; Computer science; Mathematical optimization; 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":[0.00317803,0.001471738,0.001308067,0.0008735708,0.0004617348,0.001704326,0.00156246,0.001689181,0.00143632],"category_scores_gemma":[0.01086515,0.0007839052,0.0007494043,0.0006600193,0.001753814,0.002417654,0.002519339,0.003198305,0.0003945589],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00159568,"about_ca_system_score_gemma":0.00139694,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001863095,"about_ca_topic_score_gemma":0.002435421,"domain_scores_codex":[0.99874,0.0005487522,0.00006855099,0.000183282,0.0003645417,0.00009485413],"domain_scores_gemma":[0.9970721,0.001862048,0.0002573829,0.000377498,0.00030688,0.0001241664],"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":[0.0000463813,0.00003334799,0.000663777,0.0000584411,0.00005538508,0.00008270679,0.00006135975,0.8983735,0.002323913,0.07038935,0.001771673,0.0261402],"study_design_scores_gemma":[0.000003990157,0.0000102115,0.00004518736,0.000004706022,0.00000260367,0.0000117912,0.000001932777,0.9833619,0.000290232,0.01594452,0.0003192497,0.000003705082],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.005639257,0.0001697128,0.992171,0.0003462661,0.0000336339,0.00002025688,0.00002183909,0.0002587242,0.001339223],"genre_scores_gemma":[0.6308635,0.0006431665,0.360949,0.0006070621,0.0003052607,0.000315937,0.0002145266,0.000536916,0.005564739],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.00317803,"threshold_uncertainty_score":0.01680726,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06123337170310077,"score_gpt":0.1982979884960701,"score_spread":0.1370646167929693,"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."}}