{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0004079912,0.0003647862,0.0005074999,0.0002341691,0.0001691854,0.0008843661,0.0005254286,0.0002119331,0.00008795017],"category_scores_gemma":[0.0002287064,0.0003279254,0.00004793345,0.0004171217,0.0001177348,0.0003453198,0.0009620179,0.001249075,0.000001603106],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007756336,"about_ca_system_score_gemma":0.0002806447,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008779319,"about_ca_topic_score_gemma":0.00008107511,"domain_scores_codex":[0.9978111,0.0002573118,0.0004764279,0.0007939684,0.0002845722,0.000376608],"domain_scores_gemma":[0.9987909,0.000202442,0.0003338363,0.0004647919,0.00008861218,0.0001194263],"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":[0.000004805283,0.00002919517,0.01720095,0.0003538721,0.00003016253,0.0001321478,0.00127039,0.9150462,0.000003637884,0.00150238,0.000001167686,0.0644251],"study_design_scores_gemma":[0.0003745403,0.00006392915,0.003114059,0.001129,0.00001954274,0.0001088283,0.0001055284,0.9940725,0.000005774069,0.0005836401,0.00001972204,0.0004029461],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.002086595,0.0003587691,0.9945984,0.001085471,0.0001055121,0.0002160099,0.000002056317,0.000179144,0.001367987],"genre_scores_gemma":[0.5212526,0.00008228571,0.4784469,0.00009990887,0.00001463221,0.000009862333,0.00003534701,0.00001837824,0.00004003826],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.519166,"threshold_uncertainty_score":0.9999173,"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."}}