{"id":"W3126027909","doi":"10.48550/arxiv.2101.06608","title":"Network Automatic Pruning: Start NAP and Take a Nap","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Pruning; Nap; Memory footprint; FLOPS; Machine learning; Artificial intelligence; Parallel computing","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":[0.0009431972,0.00183184,0.001130553,0.0009175123,0.0006969851,0.001276448,0.002504534,0.001433564,0.0144353],"category_scores_gemma":[0.004579104,0.0007461865,0.0008339948,0.0005909117,0.0006374053,0.002675462,0.002033989,0.002735732,0.00526573],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007330687,"about_ca_system_score_gemma":0.001321518,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004089489,"about_ca_topic_score_gemma":0.01001052,"domain_scores_codex":[0.9993394,0.00009604888,0.00003830835,0.0001361788,0.0002691754,0.0001208013],"domain_scores_gemma":[0.9987113,0.0004579007,0.00007816757,0.0003637178,0.0003167547,0.00007203146],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0009015846,0.0002403442,0.001677014,0.0004869406,0.0002097508,0.0007221845,0.0002541991,0.1048559,0.04484406,0.0321701,0.07680592,0.736832],"study_design_scores_gemma":[0.0001526756,0.0001633747,0.0009982815,0.0001277012,0.0001030676,0.0004686649,0.00009415917,0.8748907,0.04027186,0.0402057,0.04247382,0.00004990465],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.03242183,0.001286537,0.920457,0.001137377,0.0004827989,0.0003269351,0.0006735086,0.02588216,0.01733181],"genre_scores_gemma":[0.2468414,0.0008398078,0.7230934,0.001269373,0.0002237522,0.0005600473,0.002218255,0.005887251,0.01906667],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.0144353,"threshold_uncertainty_score":0.04829085,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07025394555470299,"score_gpt":0.1945067949380338,"score_spread":0.1242528493833308,"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."}}