{"id":"W4404688444","doi":"10.1109/tnnls.2024.3496492","title":"GenSafe: A Generalizable Safety Enhancer for Safe Reinforcement Learning Algorithms Based on Reduced Order Markov Decision Process Model","year":2024,"lang":"en","type":"article","venue":"IEEE Transactions on Neural Networks and Learning Systems","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"JST-Mirai Program; Natural Sciences and Engineering Research Council of Canada; Grant Foundation","keywords":"Markov decision process; Reinforcement learning; Computer science; Process (computing); Order (exchange); Machine learning; Markov process; Markov chain; Algorithm; Artificial intelligence; Mathematics; Business; Programming language; Statistics","routes":{"ca_aff":true,"ca_fund":true,"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":[0.002164274,0.001141087,0.001041621,0.0004923526,0.00036824,0.0007727322,0.001633129,0.00103809,0.002835718],"category_scores_gemma":[0.005207065,0.0005024342,0.0007451384,0.0002716895,0.001446917,0.001141961,0.002190275,0.002885742,0.0005471162],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008927947,"about_ca_system_score_gemma":0.002650965,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002781176,"about_ca_topic_score_gemma":0.003386497,"domain_scores_codex":[0.9991214,0.0002609309,0.00004649146,0.0001658004,0.0003006457,0.0001048569],"domain_scores_gemma":[0.9978098,0.001209407,0.0002537825,0.0002416599,0.0003650877,0.0001202895],"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":[0.0001301696,0.00008647829,0.00107023,0.0001557296,0.00004758903,0.0001002134,0.0001143624,0.9008088,0.004035371,0.02412719,0.001272179,0.06805177],"study_design_scores_gemma":[0.00001471627,0.0000515641,0.00004616925,0.000008675242,0.000005928511,0.00001374504,0.000003665119,0.9929825,0.0007243572,0.005600803,0.0005431767,0.000004750912],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.006403349,0.0001197189,0.9914062,0.000119313,0.0000228024,0.00005195232,0.0000362491,0.0006592455,0.001181128],"genre_scores_gemma":[0.672821,0.0003325184,0.3219301,0.0004422693,0.00006075416,0.0003798669,0.000275766,0.0003474671,0.003410399],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002835718,"threshold_uncertainty_score":0.01144594,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0171021019710014,"score_gpt":0.2699021736160693,"score_spread":0.2528000716450678,"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."}}