{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0009948565,0.0004774565,0.0004470652,0.0003401162,0.00121903,0.0008654555,0.0004265133,0.0002444986,0.00002115617],"category_scores_gemma":[0.00003514998,0.0004213924,0.0001922403,0.0007934652,0.00005111081,0.0004871572,0.000008938309,0.001323094,0.00001209885],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001822526,"about_ca_system_score_gemma":0.0001263739,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002944888,"about_ca_topic_score_gemma":0.000001832303,"domain_scores_codex":[0.9966198,0.0002663509,0.0007417957,0.0009718071,0.000670812,0.0007294873],"domain_scores_gemma":[0.9981196,0.0008092372,0.000201271,0.0004442713,0.0002179352,0.0002077117],"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":[0.000173629,0.00002412257,0.000002230149,0.0001833312,0.00005712481,0.000008067149,0.0003232532,0.963524,0.0001224632,0.0001461954,0.0002058477,0.03522969],"study_design_scores_gemma":[0.0007578419,0.0008451684,0.000002133513,0.0006644305,0.00004887924,0.00001887536,0.00006180526,0.9944227,0.0001898084,0.000006443089,0.002503985,0.0004779134],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.001049711,0.0003214628,0.9942598,0.000221761,0.002260912,0.0009769577,0.000002218709,0.0005673987,0.0003398281],"genre_scores_gemma":[0.9774342,0.0001827051,0.01275949,0.0001852258,0.0002508545,0.0002875197,0.00001734728,0.00008756954,0.00879513],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9815003,"threshold_uncertainty_score":0.9998238,"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."}}