{"id":"W4386598466","doi":"10.1109/icip49359.2023.10222624","title":"Dodging the Double Descent in Deep Neural Networks","year":2023,"lang":"en","type":"article","venue":"","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Grand Équipement National De Calcul Intensif; Canadian Institute for Advanced Research","keywords":"Regularization (linguistics); Phenomenon; Computer science; Gradient descent; Generalization; Perspective (graphical); Deep neural networks; Deep learning; Artificial neural network; Artificial intelligence; Stochastic gradient descent; Descent (aeronautics); Machine learning; Mathematical optimization; Mathematics; Epistemology; Engineering; Philosophy","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004145614,0.001324993,0.001315232,0.0007664294,0.0008026329,0.001804243,0.001561268,0.002421022,0.001850662],"category_scores_gemma":[0.01943889,0.0010223,0.0007735832,0.0005557272,0.003390536,0.004115748,0.003006996,0.004064196,0.0005669757],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001487384,"about_ca_system_score_gemma":0.001188926,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002352232,"about_ca_topic_score_gemma":0.00320419,"domain_scores_codex":[0.9981884,0.0008950458,0.00009561866,0.0002704813,0.0003812536,0.0001691577],"domain_scores_gemma":[0.9950649,0.003267671,0.0003308845,0.0006670754,0.0004073321,0.0002622152],"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":[0.0003804318,0.0001227273,0.002353914,0.00027018,0.0001290772,0.0003216561,0.0002394509,0.5712808,0.006541316,0.3360899,0.00833048,0.07394013],"study_design_scores_gemma":[0.0000194409,0.00004506504,0.0001303365,0.00003514639,0.000007452845,0.00003796909,0.00001458029,0.8745516,0.001313039,0.1220563,0.001776536,0.00001246487],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.03277612,0.001482487,0.9570457,0.003050462,0.0002260459,0.00003434692,0.00006649459,0.0006140083,0.004704258],"genre_scores_gemma":[0.6572645,0.001801234,0.3277051,0.001604441,0.0004019411,0.0002444544,0.0002741821,0.0007808669,0.009923213],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.004145614,"threshold_uncertainty_score":0.02192432,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02933795449060126,"score_gpt":0.2772926482744985,"score_spread":0.2479546937838972,"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."}}