{"id":"W3033960544","doi":"10.1109/tnnls.2021.3069970","title":"EDropout: Energy-Based Dropout and Pruning of Deep Neural Networks","year":2021,"lang":"en","type":"article","venue":"IEEE Transactions on Neural Networks and Learning Systems","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":53,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"Fujitsu","keywords":"Pruning; Artificial neural network; Dropout (neural networks); Population; Computer science; Backpropagation; Probabilistic logic; Artificial intelligence; Algorithm; Machine learning","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":[0.0002179092,0.000333382,0.0004529867,0.0001284631,0.0007240406,0.0002354145,0.0003036111,0.0001718787,0.000006030915],"category_scores_gemma":[0.000008833927,0.0003265497,0.0001227218,0.0008170086,0.0001503834,0.0003327961,0.00001525489,0.0008465471,6.539379e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000305871,"about_ca_system_score_gemma":0.00002062932,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005430487,"about_ca_topic_score_gemma":0.0000369375,"domain_scores_codex":[0.9974864,0.0004759287,0.0005281568,0.0007365874,0.0002733456,0.0004996257],"domain_scores_gemma":[0.9981511,0.0006824609,0.0002777112,0.0005151053,0.0001457381,0.0002278319],"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":[0.0000176012,0.00004204813,0.0001890853,0.00002569566,0.00002459621,0.00002097409,0.00006453234,0.9266605,0.0001786897,0.0008242537,0.00001313442,0.07193888],"study_design_scores_gemma":[0.0004544087,0.000176829,0.0001097075,0.000076085,0.00003299564,0.0001381353,0.00007849806,0.9979042,0.0001792644,0.00002079247,0.0005252029,0.0003038867],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01272422,0.00262417,0.9829543,0.0003186288,0.0008723076,0.0001965535,0.000001722242,0.0002469692,0.00006117026],"genre_scores_gemma":[0.997126,0.000214523,0.001880696,0.0002454,0.0001829805,0.00006635196,0.000006048744,0.0000414079,0.0002366007],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9844018,"threshold_uncertainty_score":0.9999186,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.009987216775506999,"score_gpt":0.2260192393803364,"score_spread":0.2160320226048293,"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."}}