{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003601197,0.0001349965,0.0001107751,0.0001051715,0.0002818463,0.000209355,0.0004923083,0.0000888504,0.00005339315],"category_scores_gemma":[0.0001275402,0.0001256959,0.00007218219,0.0001972886,0.00002065687,0.0004799314,0.0001226726,0.0003434407,0.00004027718],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003442921,"about_ca_system_score_gemma":0.00008137993,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001379146,"about_ca_topic_score_gemma":0.0000121378,"domain_scores_codex":[0.9988416,0.00001385796,0.0002383189,0.00029082,0.0002587374,0.0003566525],"domain_scores_gemma":[0.9992846,0.00006964755,0.000107306,0.0003128702,0.0001303566,0.00009528158],"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":[0.000002772848,0.000004816995,0.0002387823,0.000006912318,0.000007126034,9.675186e-8,0.0001635336,0.8490474,0.003870246,0.1436312,0.0005864137,0.002440645],"study_design_scores_gemma":[0.0002664686,0.0001438134,0.00002151609,0.000003351262,0.000004453002,0.000004018362,0.00001111191,0.9880684,0.003921894,0.001069692,0.00631552,0.000169732],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0006150215,9.791919e-7,0.9798453,0.0004050677,0.0003278475,0.0002815245,4.823078e-8,0.0004285875,0.01809558],"genre_scores_gemma":[0.6564732,0.000001428985,0.3253216,0.0002126177,0.00006314823,0.00001941819,0.000003556687,0.00001111129,0.01789386],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6558582,"threshold_uncertainty_score":0.5125731,"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."}}