{"id":"W4210647952","doi":"10.1145/3506695","title":"An Empirical Study of the Impact of Hyperparameter Tuning and Model Optimization on the Performance Properties of Deep Neural Networks","year":2022,"lang":"en","type":"article","venue":"ACM Transactions on Software Engineering and Methodology","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":198,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal; University of Alberta; Concordia University","funders":"","keywords":"Hyperparameter; Computer science; Artificial intelligence; Artificial neural network; Machine learning; Inference; Hyperparameter optimization; Pruning; Deep learning; 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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.000303055,0.00009238294,0.0001649442,0.00006758331,0.0001759509,0.000005346982,0.000365963,0.00002694887,0.000001280711],"category_scores_gemma":[0.00008105978,0.00005705259,0.00003903421,0.0003125725,0.0000535148,0.00008838165,0.00004169274,0.0002501096,7.638315e-9],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000142634,"about_ca_system_score_gemma":0.000008894861,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000009879298,"about_ca_topic_score_gemma":8.223881e-7,"domain_scores_codex":[0.99918,0.000268885,0.0001783939,0.0001714326,0.00009561511,0.0001057014],"domain_scores_gemma":[0.998641,0.0007440494,0.00007805795,0.0004792657,0.00003339138,0.00002424831],"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":[0.00002550994,0.00008189595,0.0005787665,0.000005526494,0.00001851835,5.449781e-8,0.001368022,0.9850713,0.0006353149,0.00001698926,2.091536e-7,0.01219788],"study_design_scores_gemma":[0.000129801,0.0006124711,0.00208794,0.000005055383,0.00001477099,0.0000112702,0.0001046124,0.9962534,0.0007022023,0.00002578826,3.923421e-7,0.00005224003],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4815086,0.00002925277,0.5182699,0.00003017627,0.00002696769,0.0001162221,9.924167e-7,0.00001784846,1.144098e-7],"genre_scores_gemma":[0.8189051,0.00001221424,0.1809908,0.00002236297,0.000003642977,0.00005725533,2.225695e-7,0.000007243773,0.000001244334],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.3373965,"threshold_uncertainty_score":0.2326537,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09199875935483277,"score_gpt":0.3210991654779176,"score_spread":0.2291004061230849,"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."}}