{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0008409644,0.0001223214,0.0001865052,0.0001012718,0.0001292821,0.0002071958,0.000579745,0.00004130648,0.001292315],"category_scores_gemma":[0.01604395,0.00008185102,0.0000438139,0.0002403109,0.0003493736,0.0001240588,0.00007915507,0.0001464297,0.00006073702],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002664021,"about_ca_system_score_gemma":0.00008281524,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001276955,"about_ca_topic_score_gemma":0.00001196043,"domain_scores_codex":[0.9977005,0.0001334299,0.0005871176,0.0003664614,0.001055597,0.0001569576],"domain_scores_gemma":[0.9919167,0.006188759,0.0002573399,0.0001727121,0.001372377,0.00009215467],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00068982,0.00006767394,0.00001525081,0.000003724751,0.00002080044,0.000002087957,0.0008041458,0.05348096,0.0001939721,0.8894948,0.001138819,0.05408794],"study_design_scores_gemma":[0.00003283263,0.0003258249,0.00002490265,0.000008618651,0.000002526625,2.766425e-7,0.001334247,0.5953615,0.004806139,0.3979163,0.0001295959,0.00005731996],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.001281908,0.000003684052,0.9846703,0.01093606,0.0001961481,0.0003307421,0.0009739091,0.00000885308,0.001598396],"genre_scores_gemma":[0.9631561,0.00004531416,0.03594643,0.000511345,0.00008620172,0.00003363411,0.00002741392,0.000008252024,0.0001852933],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9618742,"threshold_uncertainty_score":0.9996206,"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."}}