{"id":"W3174700868","doi":"10.1609/aaai.v35i9.16978","title":"Linearly Replaceable Filters for Deep Network Channel Pruning","year":2021,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":31,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"Ministry of Science and ICT, South Korea; Electronics and Telecommunications Research Institute","keywords":"FLOPS; Computer science; Pruning; Convolutional neural network; Deep learning; Filter (signal processing); Algorithm; Artificial intelligence; Compensation (psychology); Channel (broadcasting); Computer engineering; Pattern recognition (psychology); Parallel computing; Computer vision; Telecommunications","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":[0.0003597766,0.0002095825,0.0002549282,0.00004898628,0.0003976085,0.0002470791,0.001657618,0.00008248169,0.00002013689],"category_scores_gemma":[0.0004625689,0.0001783775,0.0001409825,0.001062265,0.0001385665,0.0004337675,0.0004716851,0.0002584019,0.00002920327],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004775678,"about_ca_system_score_gemma":0.0001023453,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000004565074,"about_ca_topic_score_gemma":0.00001121762,"domain_scores_codex":[0.9980339,0.00001443748,0.0004990406,0.0006509791,0.0003108611,0.0004907803],"domain_scores_gemma":[0.9978296,0.0002475919,0.0003845387,0.0004760882,0.0009634201,0.00009879961],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00003784627,0.00007212619,0.00002582078,0.00002959323,0.00001444119,4.71783e-7,0.0005619352,0.006634755,0.01273255,0.9487281,0.0008013353,0.03036097],"study_design_scores_gemma":[0.00002460764,0.00008576045,0.00001642905,0.0001208843,0.000008971165,0.000006294314,0.0001593623,0.4083788,0.2786322,0.3114399,0.0009504916,0.0001762656],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.008194929,0.0001005444,0.9757187,0.008698101,0.0006023198,0.000812853,0.000005392227,0.0001851267,0.005681976],"genre_scores_gemma":[0.8994915,0.00006439506,0.0981005,0.0008023871,0.0002904938,0.0001949112,0.000002122468,0.00002487612,0.001028834],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8912966,"threshold_uncertainty_score":0.7274022,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08086057145799233,"score_gpt":0.3017327517650332,"score_spread":0.2208721803070409,"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."}}