{"id":"W4293370597","doi":"10.1109/lra.2022.3196132","title":"Safe-Control-Gym: A Unified Benchmark Suite for Safe Learning-Based Control and Reinforcement Learning in Robotics","year":2022,"lang":"en","type":"article","venue":"IEEE Robotics and Automation Letters","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":47,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dynamic Systems Analysis (Canada); Vector Institute; University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada; Canada Research Chairs","keywords":"Reinforcement learning; Artificial intelligence; Computer science; Suite; Benchmark (surveying); Machine learning","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":[0.001269354,0.0003365871,0.0004465753,0.0004467249,0.0007988467,0.0003780812,0.0004400668,0.00008675001,0.00001467388],"category_scores_gemma":[0.0001828685,0.0003768486,0.00009875868,0.0004389911,0.00008986398,0.0003978628,0.0001153232,0.0006451203,0.000004103025],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002291243,"about_ca_system_score_gemma":0.0001100979,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002574526,"about_ca_topic_score_gemma":0.000004345389,"domain_scores_codex":[0.9971209,0.0003553289,0.0007372752,0.0005660666,0.0005966036,0.0006238142],"domain_scores_gemma":[0.9981421,0.000786725,0.000499915,0.0003373813,0.00008922387,0.0001446706],"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":[0.00004949594,0.00002440986,0.003199316,0.00007543294,0.00004018018,0.00001223752,0.0003985291,0.9892525,0.0014961,0.004204102,0.0002843938,0.000963274],"study_design_scores_gemma":[0.004472965,0.000475121,0.002737442,0.0000415457,0.00004359521,0.000009637048,0.00005664865,0.9893165,0.00004251023,0.00007070688,0.002316468,0.000416904],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.006357235,0.00005965916,0.9805166,0.0113608,0.0004686843,0.0009226509,0.000002667971,0.0002075701,0.0001041015],"genre_scores_gemma":[0.9711221,0.00001591353,0.0243653,0.003894972,0.00006247828,0.0001156588,0.0000467332,0.00003474267,0.0003420866],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9647649,"threshold_uncertainty_score":0.9998683,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.009230860443678211,"score_gpt":0.2213084829775076,"score_spread":0.2120776225338294,"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."}}