{"id":"W2780752111","doi":"10.1007/s10589-020-00220-z","title":"Momentum and stochastic momentum for stochastic gradient, Newton, proximal point and subspace descent methods","year":2020,"lang":"en","type":"preprint","venue":"Computational Optimization and Applications","topic":"Stochastic Gradient Optimization Techniques","field":"Computer Science","cited_by":81,"is_retracted":false,"has_abstract":false,"ca_institutions":"Université de Montréal; Mila - Quebec Artificial Intelligence Institute","funders":"","keywords":"Iterated function; Stochastic gradient descent; Momentum (technical analysis); Mathematics; Rate of convergence; Stochastic optimization; Applied mathematics; Ball (mathematics); Stochastic approximation; Mathematical optimization; Mathematical analysis; Computer science; Finance; Artificial neural network","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0003620624,0.0004925378,0.0004996904,0.0003805032,0.000475334,0.0005623339,0.0004602045,0.0002034921,0.00000562027],"category_scores_gemma":[0.0001580089,0.0005460698,0.00008144038,0.0004554356,0.0002414873,0.0002888184,0.001053603,0.0003058405,0.00000127569],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001167825,"about_ca_system_score_gemma":0.0001922975,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000006048494,"about_ca_topic_score_gemma":5.745082e-7,"domain_scores_codex":[0.9972152,0.0001078194,0.0006642455,0.001344549,0.0003243297,0.0003439032],"domain_scores_gemma":[0.9975739,0.0005624979,0.0005161942,0.0004285396,0.0005126555,0.0004062199],"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":[0.0000112693,0.00008090597,0.000005514848,0.000135982,0.00005134111,2.324287e-7,0.0004054868,0.7572226,0.00001757167,0.2392433,0.0003476399,0.002478132],"study_design_scores_gemma":[0.0006245241,0.0001027439,0.0000784949,0.00007145156,0.00008461512,0.0000216245,0.00004485656,0.8596759,0.00001999902,0.1386963,0.0001515014,0.0004279684],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.00004345071,0.0006967129,0.9871101,0.007204269,0.0001590684,0.004181574,0.0001172377,0.0004604736,0.00002710493],"genre_scores_gemma":[0.02749291,0.00009651434,0.9682357,0.0003606078,0.0001050328,0.003052742,0.000546478,0.00005614126,0.00005382013],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.1024533,"threshold_uncertainty_score":0.9996991,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03332620874769687,"score_gpt":0.320718223472419,"score_spread":0.2873920147247221,"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."}}