{"id":"W2284548247","doi":"10.1007/s10994-021-06019-1","title":"Bayesian optimization with safety constraints: safe and automatic parameter tuning in robotics","year":2021,"lang":"en","type":"preprint","venue":"Machine Learning","topic":"Gaussian Processes and Bayesian Inference","field":"Computer Science","cited_by":80,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada; Eidgenössische Technische Hochschule Zürich; Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung","keywords":"Bayesian optimization; Overshoot (microwave communication); Computer science; Robotics; Process (computing); Context (archaeology); Gaussian process; Robot; Bayesian probability; Set (abstract data type); Artificial intelligence; Probabilistic logic; Optimization problem; Mathematical optimization; Machine learning; Algorithm; Gaussian; Mathematics","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":[0.003944556,0.001411001,0.001416023,0.0009024901,0.0006999312,0.001236725,0.001523163,0.001965843,0.001108181],"category_scores_gemma":[0.01579772,0.001155897,0.000686829,0.0007688276,0.002866601,0.002376604,0.00248688,0.002664273,0.0004414776],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001202421,"about_ca_system_score_gemma":0.002542616,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003747307,"about_ca_topic_score_gemma":0.003737685,"domain_scores_codex":[0.9979292,0.0008130049,0.0001039754,0.000387686,0.0006007656,0.0001654108],"domain_scores_gemma":[0.9950672,0.003548187,0.0005015276,0.0003854292,0.0003840606,0.000113554],"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":[0.0001005408,0.00004984799,0.0007848793,0.0001116645,0.00005366311,0.00004863349,0.0001286683,0.8535184,0.004859094,0.04194605,0.0009312524,0.0974673],"study_design_scores_gemma":[0.00001749342,0.00002602109,0.0001280031,0.00002228678,0.000007071643,0.00001818289,0.000008743231,0.957024,0.002033914,0.03994678,0.0007537599,0.00001373993],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.003041705,0.0001205417,0.9961927,0.00009656699,0.00000629149,0.00001504261,0.000007740142,0.000183606,0.000335877],"genre_scores_gemma":[0.3662817,0.0005253258,0.6307365,0.000302505,0.00006553519,0.000242261,0.0001046237,0.0003556448,0.001385877],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.003944556,"threshold_uncertainty_score":0.02086109,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.008947761347103697,"score_gpt":0.2255024712557855,"score_spread":0.2165547099086818,"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."}}