{"id":"W4366327855","doi":"10.48550/arxiv.2304.08309","title":"Promises and Pitfalls of the Linearized Laplace in Bayesian Optimization","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Gaussian Processes and Bayesian Inference","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Government of Canada; Canadian Institute for Advanced Research","keywords":"Laplace's method; Laplace transform; Bayesian probability; Gaussian process; Bayesian optimization; Mathematical optimization; Maximum a posteriori estimation; Mathematics; Artificial neural network; Kernel (algebra); Applied mathematics; Covariance; Gaussian; Computer science; Artificial intelligence; Statistics; Physics; Maximum likelihood; Mathematical analysis","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001855587,0.0001744307,0.0002429309,0.0001738329,0.00006674894,0.00008827738,0.001232489,0.0001772036,0.000007020125],"category_scores_gemma":[0.00006629403,0.0001530941,0.00007056099,0.0008092392,0.0001048685,0.000226021,0.001690721,0.0003142896,0.000003579285],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004646982,"about_ca_system_score_gemma":0.0002423873,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001541482,"about_ca_topic_score_gemma":0.00009541069,"domain_scores_codex":[0.9988318,0.00009620903,0.0001947384,0.0006120453,0.0000762438,0.0001889526],"domain_scores_gemma":[0.9988553,0.00007908677,0.0002427391,0.0006758469,0.00008843865,0.00005861151],"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":[0.00002591933,0.00007157235,0.01738757,0.0003317936,0.00002994953,0.00005173183,0.0004665614,0.907479,0.00003240632,0.0737984,0.00005960267,0.0002654842],"study_design_scores_gemma":[0.0003933019,0.00002104036,0.006491241,0.0002519581,0.00001757641,0.00000169613,0.0000334471,0.9669942,0.000160232,0.02543015,0.00001322621,0.0001919689],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03999823,0.0000298857,0.9585344,0.0003615985,0.0002099186,0.0003186277,0.000007575985,0.00008684809,0.0004529421],"genre_scores_gemma":[0.9915056,0.0002094466,0.007386779,0.0000288707,0.00001400606,0.000001157728,0.000002564121,0.00001053155,0.0008410554],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9515074,"threshold_uncertainty_score":0.6242996,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05337981574930786,"score_gpt":0.1814597465117443,"score_spread":0.1280799307624364,"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."}}