{"id":"W3208296466","doi":"10.1609/aaai.v36i8.20891","title":"Convergence and Optimality of Policy Gradient Methods in Weakly Smooth Settings","year":2022,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Advanced Bandit Algorithms Research","field":"Decision Sciences","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"Vector Institute; University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Convergence (economics); Gradient method; Computer science; Mathematical optimization; Applied mathematics; Work (physics); Rate of convergence; Mathematics; Economics; Physics; Telecommunications","routes":{"ca_aff":true,"ca_fund":true,"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"],"consensus_categories":[],"category_scores_codex":[0.007144476,0.0002145217,0.0005091007,0.0006441697,0.0002910455,0.0001101122,0.002270448,0.00005930057,0.0005070998],"category_scores_gemma":[0.009762134,0.0001593962,0.0001274319,0.002935791,0.0009443412,0.0003121965,0.001485673,0.0006047828,0.00001175365],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000148238,"about_ca_system_score_gemma":0.0002888845,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003319636,"about_ca_topic_score_gemma":0.00001823495,"domain_scores_codex":[0.9954155,0.0002173798,0.001286183,0.0007427468,0.001871049,0.000467112],"domain_scores_gemma":[0.9963104,0.0012552,0.0008735281,0.000417695,0.00101166,0.0001315751],"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":[0.0005668318,0.0004801688,0.005402951,0.00007285725,0.00002138083,0.000002213333,0.00780279,0.002502391,0.1241721,0.4926808,0.0002527278,0.3660428],"study_design_scores_gemma":[0.00005902191,0.0003947311,0.003630224,0.0000652124,0.000006500173,0.00000796612,0.01056385,0.0569222,0.3719966,0.555453,0.0006692909,0.0002313489],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9585022,0.0001501185,0.02152845,0.01098066,0.0005701287,0.001293739,0.00009169479,0.00004269599,0.006840298],"genre_scores_gemma":[0.990052,0.00004814765,0.009317246,0.0001147551,0.00003034001,0.0000580424,3.4364e-7,0.00001290337,0.0003662357],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3658114,"threshold_uncertainty_score":0.9985791,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2328124462140117,"score_gpt":0.4911329585570102,"score_spread":0.2583205123429986,"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."}}