{"id":"W2099089474","doi":"","title":"PAC-Bayesian Model Selection for Reinforcement Learning","year":2010,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":21,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Reinforcement learning; Leverage (statistics); Computer science; Bayesian probability; Correctness; Selection (genetic algorithm); Artificial intelligence; Machine learning; Bayesian inference; Model selection; Algorithm","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":[0.008540191,0.002540963,0.002638661,0.001571302,0.001008047,0.003101227,0.003315815,0.002246972,0.006017182],"category_scores_gemma":[0.04576465,0.001484243,0.001299115,0.001720392,0.003698269,0.006067636,0.003609906,0.008434514,0.001565722],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.004316776,"about_ca_system_score_gemma":0.003579634,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00495114,"about_ca_topic_score_gemma":0.0043247,"domain_scores_codex":[0.9914743,0.003786962,0.0003079082,0.001047886,0.00291565,0.0004672549],"domain_scores_gemma":[0.963535,0.03018377,0.001280323,0.001998574,0.002360845,0.000641353],"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":[0.0001544577,0.00009131429,0.0005012469,0.0002639494,0.0001179644,0.0001012693,0.0001629324,0.5397377,0.0008634031,0.3920554,0.005555524,0.06039473],"study_design_scores_gemma":[0.00001210706,0.00002066613,0.00004916954,0.00002519337,0.000008643784,0.00002007943,0.000005336562,0.8285568,0.0003339677,0.1697914,0.001165516,0.0000112736],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.001039029,0.0004909728,0.9948016,0.0002874905,0.00004027741,0.00002752145,0.00004001714,0.0002384164,0.003034682],"genre_scores_gemma":[0.4904856,0.002201132,0.4962142,0.001232561,0.0006668601,0.001026685,0.0006086611,0.0007811112,0.006783193],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.008540191,"threshold_uncertainty_score":0.04516542,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01445071947155641,"score_gpt":0.2573540476256005,"score_spread":0.2429033281540441,"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."}}