{"id":"W4416046438","doi":"10.48550/arxiv.2505.21813","title":"Optimizing Data Augmentation through Bayesian Model Selection","year":2025,"lang":"en","type":"preprint","venue":"ArXiv.org","topic":"Machine Learning and Data Classification","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"","keywords":"Robustness (evolution); Model selection; Generalization; Probabilistic logic; Bayesian probability; Bayesian optimization; Marginal likelihood; Selection (genetic algorithm)","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0004495328,0.0002531666,0.0002265027,0.0001346837,0.0002308903,0.0002929957,0.002408842,0.0002210806,0.00001247899],"category_scores_gemma":[0.0001250319,0.0002701159,0.0000549882,0.0003350219,0.00002466851,0.001093062,0.002912177,0.0007423622,0.00005897474],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001280553,"about_ca_system_score_gemma":0.0003406024,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003346458,"about_ca_topic_score_gemma":0.0000487313,"domain_scores_codex":[0.9976932,0.000157608,0.0003724919,0.00124915,0.0002797355,0.0002477964],"domain_scores_gemma":[0.9970873,0.00006602818,0.0003083927,0.002386485,0.00009955032,0.00005224609],"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":[0.00003698706,0.0003280432,0.0917716,0.0006429382,0.0002496729,0.000006518176,0.004511831,0.777386,0.001362769,0.03621221,0.02681164,0.06067986],"study_design_scores_gemma":[0.0001544962,0.00001189639,0.004076965,0.0001003468,0.00003487644,0.000001743125,0.00001863877,0.9902374,0.0002219805,0.00205882,0.002834372,0.0002484241],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.003679507,0.0001392482,0.9876393,0.002593082,0.0008385283,0.0002539396,0.00006050095,0.0004833227,0.004312542],"genre_scores_gemma":[0.45427,0.0002648686,0.5386168,0.0007014448,0.0002270161,0.00004896434,0.002968335,0.00002083176,0.002881781],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.4505905,"threshold_uncertainty_score":0.9999751,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1231120383278702,"score_gpt":0.3565369785081619,"score_spread":0.2334249401802917,"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."}}