{"id":"W4288598872","doi":"10.48550/arxiv.1902.08673","title":"Ising-Dropout: A Regularization Method for Training and Compression of\\n Deep Neural Networks","year":2019,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Machine Learning and Data Classification","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"MNIST database; Overfitting; Computer science; Initialization; Artificial intelligence; Regularization (linguistics); Dropout (neural networks); Inference; Artificial neural network; Machine learning; Deep neural networks; Deep 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":[],"consensus_categories":[],"category_scores_codex":[0.0004313038,0.0001824105,0.0002975994,0.000169544,0.0001253799,0.00009265956,0.0006934219,0.0002258571,0.000002843939],"category_scores_gemma":[0.0000679569,0.0001987328,0.00009265714,0.0002474946,0.00004716575,0.0002625921,0.0007436778,0.0003329575,9.066668e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003603995,"about_ca_system_score_gemma":0.00003984021,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004239487,"about_ca_topic_score_gemma":0.000006508421,"domain_scores_codex":[0.9985603,0.0002384792,0.0001913958,0.0007637503,0.00005907619,0.0001870169],"domain_scores_gemma":[0.9983665,0.000273936,0.0004162554,0.000750149,0.000121395,0.00007170733],"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":[0.00002779958,0.00002193723,0.001336406,0.0001079791,0.00002551007,0.000002756898,0.0003859893,0.8588201,0.00008041437,0.09814794,0.00002042311,0.04102269],"study_design_scores_gemma":[0.0003861921,0.00005142246,0.001080429,0.00006761813,0.00004688873,0.000002886146,0.00005616708,0.9883907,0.00002331036,0.009534048,0.0001645167,0.0001957622],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.006939852,0.00008639367,0.9917872,0.0001108615,0.0003512157,0.0003163615,0.000008335231,0.0001006585,0.0002991424],"genre_scores_gemma":[0.9403412,0.00002968621,0.05918189,0.00003883555,0.00004812098,8.754245e-7,0.0001204466,0.00001278941,0.0002261068],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9334014,"threshold_uncertainty_score":0.8104088,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0765926445852397,"score_gpt":0.2345336015730993,"score_spread":0.1579409569878596,"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."}}