{"id":"W2970615870","doi":"10.48550/arxiv.1911.00937","title":"Preventing Gradient Attenuation in Lipschitz Constrained Convolutional Networks","year":2019,"lang":"en","type":"article","venue":"arXiv (Cornell University)","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":33,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Lipschitz continuity; Robustness (evolution); Computer science; Norm (philosophy); Mathematical optimization; Convolution (computer science); Algorithm; Mathematics; Artificial intelligence; Artificial neural network; Pure 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003177044,0.001494582,0.001066533,0.0005705413,0.000668143,0.001322549,0.001717007,0.001738305,0.002320702],"category_scores_gemma":[0.02180559,0.0008546428,0.0008217758,0.0004827614,0.002906682,0.003958192,0.005353268,0.003925668,0.0005815801],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001501997,"about_ca_system_score_gemma":0.001248084,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001454077,"about_ca_topic_score_gemma":0.001573371,"domain_scores_codex":[0.9974909,0.0009002296,0.0001356684,0.000409877,0.0008094787,0.0002539541],"domain_scores_gemma":[0.9925146,0.004710585,0.0006056422,0.001435167,0.0005186948,0.0002152725],"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":[0.0003552053,0.00007924472,0.001234212,0.0001942367,0.00005548918,0.0002438857,0.0001935701,0.7552347,0.01637085,0.1646622,0.003605148,0.05777123],"study_design_scores_gemma":[0.00001960043,0.00005882998,0.0001291397,0.00002491313,0.000008273149,0.00005363994,0.00001415915,0.9238441,0.007049003,0.06733013,0.001456676,0.00001151034],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.034575,0.0002007823,0.9597374,0.0005397819,0.00003502886,0.00004886707,0.00008850675,0.001141166,0.003633495],"genre_scores_gemma":[0.8429547,0.0004475342,0.1503093,0.0006166655,0.00008491714,0.0002920559,0.000376014,0.0005925411,0.004326328],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003177044,"threshold_uncertainty_score":0.01680201,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02843321358506987,"score_gpt":0.1780032827735786,"score_spread":0.1495700691885087,"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."}}