{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003263012,0.001746168,0.001839419,0.001370671,0.0008950364,0.002364419,0.001639939,0.002990724,0.003983271],"category_scores_gemma":[0.01592976,0.0007475049,0.001011235,0.002890756,0.00302029,0.003639078,0.002345306,0.004955544,0.001435748],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001375346,"about_ca_system_score_gemma":0.002644965,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004386762,"about_ca_topic_score_gemma":0.004196927,"domain_scores_codex":[0.9982734,0.0008129227,0.00009223101,0.0002154866,0.0005163687,0.00008958897],"domain_scores_gemma":[0.9974171,0.001447865,0.0002017204,0.0002720524,0.0004992285,0.0001619749],"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":[0.0001077612,0.00009457194,0.0003872,0.0002940607,0.00007514485,0.00008530503,0.0001358386,0.1601732,0.001723599,0.7488347,0.0122076,0.07588104],"study_design_scores_gemma":[0.00001317172,0.00002685353,0.0001905938,0.00003322356,0.00001765017,0.00003627708,0.0000154061,0.7362049,0.0005806736,0.2555812,0.00727763,0.00002249134],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.00281172,0.002533095,0.9902812,0.0006309514,0.0004234188,0.00002820845,0.00006935365,0.0002205072,0.003001621],"genre_scores_gemma":[0.2095441,0.008262073,0.7348788,0.0007448133,0.002493819,0.0006131019,0.0006736057,0.00135728,0.04143241],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.004386762,"threshold_uncertainty_score":0.01725668,"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."}}