{"id":"W3158392704","doi":"","title":"On the Suboptimality of Negative Momentum for Minimax Optimization","year":2020,"lang":"en","type":"article","venue":"International Conference on Artificial Intelligence and Statistics","topic":"Advanced Bandit Algorithms Research","field":"Decision Sciences","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Momentum (technical analysis); Minimax; Convergence (economics); Mathematical optimization; Mathematics; Rate of convergence; Simple (philosophy); Applied mathematics; Computer science; Economics; Key (lock)","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":[0.006444142,0.00169606,0.001550065,0.001088761,0.00124514,0.002185242,0.001363131,0.001765114,0.003469512],"category_scores_gemma":[0.03999656,0.0007333848,0.0009708465,0.0006415041,0.005113989,0.003784037,0.003094444,0.004169783,0.0005661581],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001781007,"about_ca_system_score_gemma":0.002563095,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00295791,"about_ca_topic_score_gemma":0.002119435,"domain_scores_codex":[0.9977739,0.001160042,0.00007541524,0.0002920106,0.0004153981,0.0002832691],"domain_scores_gemma":[0.9821675,0.01477011,0.0007654498,0.0008063582,0.0009552345,0.0005352302],"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":[0.0002542959,0.00008810157,0.001205418,0.0002636026,0.00007224459,0.0002614163,0.0002524696,0.3905961,0.002904053,0.5795569,0.00410141,0.02044401],"study_design_scores_gemma":[0.00002001389,0.00005611551,0.0001494408,0.00005985485,0.000009370485,0.00004397946,0.0000262707,0.8852212,0.0005626499,0.1129106,0.0009247198,0.00001571732],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02860283,0.00125388,0.9528089,0.001853927,0.0002133674,0.0000784881,0.00005266267,0.0002241879,0.01491177],"genre_scores_gemma":[0.8377011,0.001785228,0.149107,0.0011138,0.0002325383,0.0003726078,0.0001056043,0.0004374417,0.00914468],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006444142,"threshold_uncertainty_score":0.03408027,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.4766689215919608,"score_gpt":0.4878975526146724,"score_spread":0.01122863102271154,"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."}}