{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004297545,0.001944089,0.002660741,0.0006502641,0.0005695617,0.001616885,0.002210754,0.002265602,0.003234463],"category_scores_gemma":[0.01923783,0.001187303,0.0009558189,0.0006457453,0.002643615,0.002936811,0.002687052,0.004106457,0.000481903],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002921406,"about_ca_system_score_gemma":0.002061725,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006532812,"about_ca_topic_score_gemma":0.003883637,"domain_scores_codex":[0.9983766,0.000770126,0.00006622304,0.0003562263,0.0002129473,0.0002179875],"domain_scores_gemma":[0.9869078,0.01099505,0.0008075689,0.0004361706,0.0004248671,0.0004286142],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0001287683,0.00006390307,0.0005349607,0.00007854681,0.00005560245,0.00005733547,0.0000489436,0.9600255,0.0003948712,0.03024451,0.0005600091,0.007806947],"study_design_scores_gemma":[0.00001182887,0.00001627415,0.00003329765,0.00000484093,0.000003930892,0.000003476604,0.000002891682,0.9901966,0.0000870822,0.009557727,0.00007919826,0.000002889061],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04323617,0.0009295819,0.9511062,0.0009550775,0.00006668054,0.00007623476,0.00009655034,0.0003575801,0.003175835],"genre_scores_gemma":[0.8814241,0.0006222217,0.1108362,0.0005068816,0.0001162499,0.0002732524,0.0002337445,0.0001928467,0.005794432],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006532812,"threshold_uncertainty_score":0.02272785,"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."}}