{"id":"W2898859254","doi":"10.48550/arxiv.1810.12805","title":"Piecewise Strong Convexity of Neural Networks","year":2018,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Stochastic Gradient Optimization Techniques","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Maxima and minima; Piecewise; Mathematics; Convexity; Artificial neural network; Differentiable function; Stochastic gradient descent; Norm (philosophy); Applied mathematics; Regularization (linguistics); Open set; Convex function; Regular polygon; Gradient descent; Mathematical optimization; Mathematical analysis; Computer science; Pure mathematics; Artificial intelligence; Geometry","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":[0.000210817,0.0002747384,0.0003535012,0.0002547173,0.00008634735,0.00005925059,0.002094486,0.0002835847,0.00003964592],"category_scores_gemma":[0.00004401598,0.0003268635,0.0001804409,0.000545754,0.0003021819,0.0002811386,0.002332918,0.0004049156,0.000007887365],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001097639,"about_ca_system_score_gemma":0.00009580335,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006601235,"about_ca_topic_score_gemma":0.000006008912,"domain_scores_codex":[0.9983626,0.0001087869,0.0002574417,0.0008584525,0.00009772598,0.0003149735],"domain_scores_gemma":[0.9976152,0.00009482959,0.0004388634,0.001367946,0.0003421799,0.0001410486],"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":[0.00001848535,0.00007305491,0.002232808,0.00003819959,0.00005710368,0.00003542081,0.0001000338,0.7625285,0.000002688143,0.2339612,0.0005019031,0.0004506445],"study_design_scores_gemma":[0.0002130008,0.00008497881,0.0004351929,0.0000541472,0.00003633466,0.000002275252,0.000009975634,0.9763373,0.0001092485,0.02243682,0.00001373913,0.0002669916],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02636324,0.00003351388,0.9706501,0.0000298504,0.0007704597,0.0003137335,0.000009805373,0.0004127122,0.001416522],"genre_scores_gemma":[0.9859583,0.00003067622,0.01368715,0.00004104339,0.00006988084,0.000001024167,0.0000125142,0.00001513015,0.0001843009],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.959595,"threshold_uncertainty_score":0.9999183,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07283423281946094,"score_gpt":0.1943522013176432,"score_spread":0.1215179684981822,"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."}}