{"id":"W3175841314","doi":"","title":"Nearly Minimax Optimal Reinforcement Learning for Linear Mixture MDPs","year":2021,"lang":"en","type":"article","venue":"Conference on Learning Theory","topic":"Advanced Bandit Algorithms Research","field":"Decision Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Reinforcement learning; Markov decision process; Mathematics; Minimax; Logarithm; Estimator; Bounded function; Combinatorics; Discrete mathematics; Regret; Mathematical optimization; Markov process; Computer science; Artificial intelligence; Statistics","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","metaepi_narrow","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.005012404,0.0003266715,0.0005150852,0.0003043687,0.0007814612,0.0005602275,0.0009640079,0.0002190779,0.003917007],"category_scores_gemma":[0.02804554,0.0002704098,0.0002631378,0.000725395,0.0002924948,0.0003775758,0.000385049,0.001400835,0.001219223],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009700968,"about_ca_system_score_gemma":0.0005536968,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000002503219,"about_ca_topic_score_gemma":0.000001937233,"domain_scores_codex":[0.9947803,0.0009532981,0.000679712,0.001069474,0.001713616,0.0008036008],"domain_scores_gemma":[0.9916461,0.00530716,0.0003388671,0.0007551978,0.001674466,0.0002782703],"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":[0.001229067,0.0001244989,0.0003676342,0.00003343076,0.00009495215,0.0001468932,0.003442448,0.6244938,0.004454202,0.05522158,0.002233653,0.3081579],"study_design_scores_gemma":[0.002226338,0.001866974,0.0005547047,0.0002004954,0.00003248378,0.00003954548,0.01101002,0.5287054,0.01386114,0.05144144,0.3891751,0.0008863292],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02360537,0.0002567903,0.9585086,0.001306941,0.0004704106,0.0004419152,0.000006517675,0.0001922457,0.0152112],"genre_scores_gemma":[0.7834133,0.00008969069,0.01105389,0.0002424352,0.0003878102,0.00009086706,0.000060687,0.00006121117,0.2046001],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9474548,"threshold_uncertainty_score":0.9999748,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1281220911290732,"score_gpt":0.423888662028909,"score_spread":0.2957665708998357,"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."}}