{"id":"W2739747865","doi":"10.24963/ijcai.2017/290","title":"Constrained Bayesian Reinforcement Learning via Approximate Linear Programming","year":2017,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo; Kootenay Association for Science & Technology","funders":"Institute for Information and Communications Technology Promotion; Defense Acquisition Program Administration; Korea Advanced Institute of Science and Technology; Agency for Defense Development; Ministry of Science, ICT and Future Planning","keywords":"Reinforcement learning; Computer science; Artificial intelligence; Bayesian probability; Mathematical optimization; Machine learning; Linear programming; Bayesian optimization; State (computer science); Algorithm; 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":"codex-gemma-dda1882f352a","candidate_categories":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0005992876,0.0002408588,0.0002324313,0.0000981561,0.001213396,0.001069936,0.001869131,0.00009156088,0.00009709175],"category_scores_gemma":[0.0002512796,0.0002183669,0.0001004577,0.00009931535,0.0001745618,0.0009855592,0.0008294573,0.0003612784,0.0001603084],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005242438,"about_ca_system_score_gemma":0.00006992065,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005835887,"about_ca_topic_score_gemma":0.000002077435,"domain_scores_codex":[0.997987,0.00004624168,0.0004095842,0.0004515508,0.0004855093,0.0006200809],"domain_scores_gemma":[0.9978651,0.000052952,0.0004307569,0.001354315,0.0001144499,0.0001824616],"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":[0.00000765384,0.00002519944,0.003250204,0.00005519732,0.00006326764,0.00004405678,0.0006145881,0.8323537,0.0004088587,0.07938593,0.00009264463,0.08369869],"study_design_scores_gemma":[0.0004715454,0.0001758174,0.0002339013,0.00003206717,0.000007333368,0.0000180718,0.00004463903,0.9887114,0.0009744252,0.0001257043,0.008914045,0.0002910348],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0001104652,0.000005850897,0.9358572,0.0004340626,0.0002972036,0.0003376806,4.336112e-8,0.0005640608,0.06239338],"genre_scores_gemma":[0.7201626,0.000005201874,0.2729431,0.000121291,0.00007731198,0.00002000328,0.000004420723,0.00001655393,0.006649508],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7200521,"threshold_uncertainty_score":0.999967,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0208781681538385,"score_gpt":0.2726533437933096,"score_spread":0.2517751756394711,"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."}}