{"id":"W3180051575","doi":"10.1109/access.2021.3108545","title":"Taxonomy of Saliency Metrics for Channel Pruning","year":2021,"lang":"en","type":"preprint","venue":"IEEE Access","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Trinity College","funders":"Science Foundation Ireland","keywords":"Pruning; Computer science; Metric (unit); Artificial intelligence; Convolutional neural network; Machine learning; Taxonomy (biology); Reduction (mathematics); Context (archaeology); Computation; Pattern recognition (psychology); Algorithm; Mathematics","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0007311697,0.0003166101,0.0006163213,0.0004740171,0.0001355642,0.0005870464,0.004767936,0.0003003595,0.00000995911],"category_scores_gemma":[0.0007049232,0.000341408,0.0002618621,0.0009275414,0.0000528222,0.0008673701,0.004053413,0.0006999429,0.000001442039],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009913276,"about_ca_system_score_gemma":0.0004536553,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001892445,"about_ca_topic_score_gemma":0.00001129028,"domain_scores_codex":[0.9975741,0.0001309379,0.0005608346,0.0008976074,0.0004439304,0.0003926132],"domain_scores_gemma":[0.9968206,0.0005069904,0.0008316283,0.001264075,0.0004874424,0.00008920862],"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":[0.00001594065,0.0001601705,0.002135162,0.00141929,0.0002094858,0.0000335635,0.001438439,0.935495,0.0001466945,0.006693762,0.000925338,0.05132717],"study_design_scores_gemma":[0.0006624231,0.00006142718,0.0006574549,0.0005074507,0.0000876757,0.000007319102,0.000097934,0.9782059,0.008923003,0.007037254,0.002875635,0.0008765276],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.00745284,0.0003593811,0.9847741,0.0001726085,0.005274242,0.0007826337,0.00001025186,0.000148101,0.001025834],"genre_scores_gemma":[0.7740569,0.00003385438,0.2249399,0.000102373,0.0004142566,0.0003157761,0.00002111376,0.000033543,0.00008234938],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.766604,"threshold_uncertainty_score":0.9999038,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1039471832573742,"score_gpt":0.3452101734254416,"score_spread":0.2412629901680674,"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."}}