{"id":"W2166107799","doi":"10.1162/089976600300015187","title":"Gradient-Based Optimization of Hyperparameters","year":2000,"lang":"en","type":"article","venue":"Neural Computation","topic":"Sparse and Compressive Sensing Techniques","field":"Engineering","cited_by":478,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Hyperparameter; Cholesky decomposition; Gradient descent; Hyperparameter optimization; Computer science; Selection (genetic algorithm); Mathematics; Mathematical optimization; Model selection; Applied mathematics; Computation; Quadratic equation; Artificial intelligence; Algorithm; Machine learning; Artificial neural network; Support vector machine","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":[0.002636237,0.001836448,0.001340689,0.0008462488,0.0005776387,0.001361608,0.001160384,0.001718134,0.002688183],"category_scores_gemma":[0.01129065,0.0008774374,0.0005642779,0.0007312057,0.001582157,0.002340184,0.001331006,0.002090892,0.001456302],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009447302,"about_ca_system_score_gemma":0.00151671,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001651361,"about_ca_topic_score_gemma":0.001877019,"domain_scores_codex":[0.999025,0.0004841719,0.00005284218,0.0001567294,0.0002163507,0.00006483479],"domain_scores_gemma":[0.9982954,0.001033631,0.0001506405,0.0001946693,0.0002827075,0.00004289705],"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":[0.0000896578,0.00006528314,0.0005526655,0.0001658389,0.00007661761,0.00004750532,0.00007980233,0.8427498,0.006531926,0.04337621,0.003016669,0.1032481],"study_design_scores_gemma":[0.00002162737,0.00003042166,0.0001508543,0.00002778442,0.0000107483,0.00002251599,0.00001047419,0.9744647,0.001988996,0.02170081,0.001556693,0.00001432873],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.004989334,0.000215894,0.9927465,0.0001999,0.00002602489,0.0000443929,0.00003169875,0.0003171736,0.001429005],"genre_scores_gemma":[0.3203906,0.0007433142,0.6721972,0.0004090204,0.000156526,0.0007232433,0.0003121126,0.0006893451,0.004378588],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.002688183,"threshold_uncertainty_score":0.01394194,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01406182111473949,"score_gpt":0.2205749250895469,"score_spread":0.2065131039748074,"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."}}