{"id":"W4327559110","doi":"10.1016/j.cviu.2023.103682","title":"Grow-push-prune: Aligning deep discriminants for effective structural network compression","year":2023,"lang":"en","type":"article","venue":"Computer Vision and Image Understanding","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":false,"ca_institutions":"McGill University","funders":"","keywords":"Computer science; MNIST database; Pruning; Artificial intelligence; Deep learning; Machine learning; Construct (python library); Residual; Set (abstract data type); Algorithm","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":[0.0002413564,0.0002478052,0.00026875,0.0001424934,0.001038203,0.0003962114,0.0004654988,0.00006443222,0.000002076292],"category_scores_gemma":[0.00001649267,0.0002074176,0.00008564018,0.0006845181,0.0001041542,0.0008736269,0.0007633728,0.0001477495,0.00001077021],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001041799,"about_ca_system_score_gemma":0.000009335549,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000001351235,"about_ca_topic_score_gemma":0.000002101693,"domain_scores_codex":[0.9982308,0.00008239076,0.0002646943,0.0006725949,0.0002082007,0.0005413013],"domain_scores_gemma":[0.9985474,0.0007372338,0.0001318104,0.0003745304,0.00005231202,0.0001566925],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001544514,0.00005248228,0.0009033966,0.0002602783,0.00009252632,0.00007730508,0.003202376,0.01670882,0.01332334,0.39451,0.0424764,0.5282386],"study_design_scores_gemma":[0.0006762313,0.0001623868,0.002589715,0.0001431531,0.00000976625,0.00002183413,0.00008952726,0.8967773,0.0003938456,0.09827454,0.0005890648,0.0002726133],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.005812129,0.000120315,0.9913408,0.0006751477,0.0007377428,0.0007340779,0.000003083385,0.0004808028,0.00009593109],"genre_scores_gemma":[0.77569,0.00004830636,0.223513,0.0002749122,0.0003383942,0.00005358164,0.00002394793,0.0000277301,0.00003013229],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8800685,"threshold_uncertainty_score":0.8458246,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0348156777220367,"score_gpt":0.3205194251396037,"score_spread":0.285703747417567,"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."}}