{"id":"W3023384417","doi":"10.1109/isit45174.2021.9518176","title":"Regret Bounds for Safe Gaussian Process Bandit Optimization","year":2021,"lang":"en","type":"preprint","venue":"","topic":"Advanced Bandit Algorithms Research","field":"Decision Sciences","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Regret; Mathematical optimization; Bayesian optimization; Computer science; Stochastic game; Gaussian process; Set (abstract data type); Constraint (computer-aided design); Function (biology); Bayesian probability; Gaussian; Mathematics; Artificial intelligence; Machine learning; Mathematical economics","routes":{"ca_aff":true,"ca_fund":false,"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":[0.008742636,0.00216435,0.002076475,0.001145666,0.001233294,0.003255131,0.002337099,0.002499236,0.004917461],"category_scores_gemma":[0.04400567,0.0007757812,0.001126684,0.001265151,0.004184685,0.003258207,0.003536073,0.005574663,0.001169255],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00351853,"about_ca_system_score_gemma":0.002701814,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005050872,"about_ca_topic_score_gemma":0.003400249,"domain_scores_codex":[0.9960521,0.001870209,0.0001281652,0.0003904209,0.001095206,0.0004639554],"domain_scores_gemma":[0.9655074,0.02951311,0.001348521,0.001321598,0.00169152,0.0006179883],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0001558859,0.00005353627,0.0005751355,0.0001295653,0.00004568518,0.00005327226,0.00007851737,0.8787871,0.0003929981,0.1015497,0.002216661,0.01596198],"study_design_scores_gemma":[0.00000907281,0.00001722162,0.00007528962,0.00002736921,0.000007030515,0.00001029712,0.00001080669,0.9585309,0.0002189749,0.04073975,0.0003481495,0.000005090783],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01307157,0.002096569,0.9718916,0.001492355,0.00009779086,0.00006611341,0.0001561216,0.0004893536,0.01063856],"genre_scores_gemma":[0.7280462,0.003415161,0.2539479,0.001548373,0.0004484654,0.0006141427,0.0007750442,0.0008727546,0.01033201],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008742636,"threshold_uncertainty_score":0.04623598,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1769952684360234,"score_gpt":0.4986446823155516,"score_spread":0.3216494138795282,"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."}}