{"id":"W2963582321","doi":"","title":"Unifying PAC and Regret: Uniform PAC Bounds for Episodic Reinforcement Learning","year":2017,"lang":"en","type":"article","venue":"Neural Information Processing Systems","topic":"Advanced Bandit Algorithms Research","field":"Decision Sciences","cited_by":103,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Regret; Reinforcement learning; Computer science; Bridge (graph theory); State (computer science); Mathematical optimization; Algorithm; Artificial intelligence; Machine learning; Mathematics","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":[0.01294393,0.002299588,0.0024161,0.001442923,0.001011621,0.004188729,0.003220331,0.002530761,0.003264911],"category_scores_gemma":[0.08591015,0.0009516589,0.001255024,0.001388084,0.00545291,0.01021006,0.004247387,0.005985639,0.0004688753],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003091693,"about_ca_system_score_gemma":0.002774602,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002032767,"about_ca_topic_score_gemma":0.001396243,"domain_scores_codex":[0.9904659,0.003769127,0.0004711986,0.00161483,0.002811109,0.0008678396],"domain_scores_gemma":[0.9264156,0.05962985,0.00407852,0.005861727,0.00291434,0.001099981],"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":[0.0002116735,0.0001109223,0.001386446,0.0002055274,0.0001124877,0.00009542967,0.000139257,0.6219187,0.0009638459,0.3346011,0.001608242,0.03864648],"study_design_scores_gemma":[0.00001390119,0.00008531291,0.0003079187,0.00005590195,0.00001972351,0.00004476952,0.00001783077,0.8218678,0.00108898,0.1757448,0.0007338598,0.00001929725],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.007185695,0.0008937417,0.9867079,0.0004496643,0.00007076332,0.00004638979,0.00006451534,0.0002299015,0.004351404],"genre_scores_gemma":[0.7538124,0.001631089,0.2392943,0.0008435114,0.0005349756,0.0004809819,0.0002589545,0.0004568239,0.00268703],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01294393,"threshold_uncertainty_score":0.0684548,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1348757675199226,"score_gpt":0.4272213443219531,"score_spread":0.2923455768020305,"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."}}