{"id":"W3131910208","doi":"10.3390/jimaging8030064","title":"Rethinking Weight Decay for Efficient Neural Network Pruning","year":2022,"lang":"en","type":"article","venue":"Journal of Imaging","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":27,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Grand Équipement National De Calcul Intensif; Association Nationale de la Recherche et de la Technologie; Compute Canada","keywords":"Pruning; Computer science; Scalability; Generalization; Artificial neural network; Smoothing; Artificial intelligence; Deep neural networks; Machine learning; Face (sociological concept); Field (mathematics); Mathematics","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":[0.0008446135,0.0001007169,0.0001638174,0.0001082665,0.0007482654,0.00009689264,0.0009744885,0.000008965636,0.00000626808],"category_scores_gemma":[0.00003036698,0.00009511322,0.0001306758,0.0005383203,0.00002301472,0.0003254243,0.0004015196,0.0004211122,7.159197e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001275809,"about_ca_system_score_gemma":0.00006208479,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":4.984692e-7,"about_ca_topic_score_gemma":2.119035e-7,"domain_scores_codex":[0.998611,0.00008122342,0.0004159331,0.0001876609,0.0003697833,0.0003344537],"domain_scores_gemma":[0.9986517,0.0003493736,0.0005526912,0.0002598389,0.0001038473,0.00008258536],"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":[0.00001412216,0.00002688558,0.000462645,0.000003415552,0.00001021918,0.00003414091,0.0005972947,0.9427998,0.001063239,0.01227924,0.003506378,0.03920261],"study_design_scores_gemma":[0.0003727651,0.00005369661,0.0001739334,0.00002300166,0.00001209001,0.0006204821,0.00003777646,0.9522783,0.0003947653,0.0167832,0.02911958,0.0001304783],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01582595,0.00110712,0.9758326,0.005640917,0.001176682,0.0001633532,6.394895e-7,0.00005287119,0.0001998429],"genre_scores_gemma":[0.7748762,0.000006617945,0.2235461,0.0009609912,0.0005445816,0.00001892609,5.586187e-7,0.00001517949,0.00003083035],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7590503,"threshold_uncertainty_score":0.5755128,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01754629696052764,"score_gpt":0.2693144594109994,"score_spread":0.2517681624504717,"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."}}