{"id":"W1973749650","doi":"10.1016/j.artint.2012.04.006","title":"Reinforcement learning with limited reinforcement: Using Bayes risk for active learning in POMDPs","year":2012,"lang":"en","type":"article","venue":"Artificial Intelligence","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":34,"is_retracted":false,"has_abstract":false,"ca_institutions":"McGill University","funders":"","keywords":"Partially observable Markov decision process; Reinforcement learning; Computer science; Markov decision process; Task (project management); Artificial intelligence; Machine learning; Action (physics); Variety (cybernetics); Bayes' theorem; Bayesian probability; Markov chain; Markov process; Markov model; 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":[0.006558639,0.001313484,0.002881939,0.0007208174,0.0006320507,0.001717184,0.002713748,0.002118117,0.002852238],"category_scores_gemma":[0.01993205,0.001326046,0.00101635,0.0005020507,0.002077353,0.003857485,0.002919303,0.003142472,0.0003358923],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001318277,"about_ca_system_score_gemma":0.001741982,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004967673,"about_ca_topic_score_gemma":0.003661803,"domain_scores_codex":[0.9979107,0.001103418,0.0001309064,0.0002829829,0.0004084478,0.0001636363],"domain_scores_gemma":[0.9871677,0.01080015,0.0004832625,0.0004155688,0.0007593548,0.0003740038],"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":[0.0001658692,0.00007727094,0.0005300401,0.000103679,0.00005945103,0.00004018589,0.00006454234,0.9522944,0.000323611,0.0191981,0.0003936534,0.02674912],"study_design_scores_gemma":[0.00001094947,0.00001283797,0.00001510849,0.000005709424,0.000004304819,0.000002302532,0.00000173495,0.9937317,0.00007244762,0.006100242,0.00004014709,0.000002630339],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01223663,0.0002332744,0.9862854,0.0001744185,0.00003791055,0.0000362258,0.0000209342,0.0001695419,0.0008055962],"genre_scores_gemma":[0.8339967,0.0002611059,0.1629782,0.0001694594,0.0000976673,0.0003314084,0.00008822446,0.0001223179,0.00195496],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006558639,"threshold_uncertainty_score":0.03468585,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06131471456678903,"score_gpt":0.3031917075469479,"score_spread":0.2418769929801589,"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."}}