{"id":"W2951603627","doi":"","title":"On the Number of Linear Regions of Deep Neural Networks","year":2014,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Neural Networks and Applications","field":"Computer Science","cited_by":488,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canadian Institute for Advanced Research; Université de Montréal","funders":"","keywords":"Rectifier (neural networks); Piecewise linear function; Computation; Artificial neural network; Computer science; Feedforward neural network; Feed forward; Deep learning; Activation function; Homogeneous space; Piecewise; Algorithm; Artificial intelligence; Mathematics; Stochastic neural network; Recurrent neural network; 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":[],"consensus_categories":[],"category_scores_codex":[0.0001396695,0.0001698645,0.0002423423,0.00004842724,0.00011792,0.00001901425,0.001764987,0.000144344,0.00002155014],"category_scores_gemma":[0.0000151136,0.0001390199,0.0002118215,0.0004543238,0.0001591811,0.00006302528,0.000977441,0.0004540488,0.0000139197],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001940985,"about_ca_system_score_gemma":0.00002040006,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004967698,"about_ca_topic_score_gemma":0.00001176568,"domain_scores_codex":[0.9989624,0.0001187398,0.0001904328,0.0004732802,0.00006716325,0.000187988],"domain_scores_gemma":[0.9976642,0.0003893278,0.000348345,0.00140895,0.0001214919,0.0000676962],"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":[0.000004156348,0.00002775317,0.0002242523,0.000006671767,0.00001421376,0.000002886441,0.00001673964,0.5188372,0.000001993063,0.480374,0.0002734051,0.0002166561],"study_design_scores_gemma":[0.00008800693,0.00001909985,0.0002777513,0.00003093658,0.00002415208,0.000001648205,0.000007026317,0.9330857,0.00002928446,0.06621289,0.0001028191,0.000120695],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1757635,0.0000127768,0.8214738,0.0004185324,0.0001957598,0.0001899341,0.000003207947,0.00004911238,0.001893294],"genre_scores_gemma":[0.9989383,0.00004921901,0.0005177889,0.00018432,0.00006362818,0.000001225189,0.000004301885,0.000009311565,0.0002319593],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8231747,"threshold_uncertainty_score":0.5669068,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05273342306533113,"score_gpt":0.198821311659622,"score_spread":0.1460878885942908,"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."}}