{"id":"W3122551441","doi":"10.5281/zenodo.4532597","title":"NNGeometry: Easy and Fast Fisher Information Matrices and Neural Tangent Kernels in PyTorch","year":2021,"lang":"en","type":"article","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Neural Networks and Applications","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Artificial neural network; Linear algebra; Computer science; Fisher information; Tangent; TRACE (psycholinguistics); Matrix norm; Matrix (chemical analysis); Cardinality (data modeling); Theoretical computer science; Algebra over a field; Algorithm; Artificial intelligence; Mathematics; Machine learning; Data mining; Pure mathematics; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001542813,0.001921362,0.001521762,0.001416403,0.0006999316,0.002419431,0.00468348,0.001796515,0.1113579],"category_scores_gemma":[0.009886682,0.0019099,0.001595841,0.001260103,0.0009122773,0.004783238,0.004799948,0.003814385,0.07432261],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001154133,"about_ca_system_score_gemma":0.001503298,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006249676,"about_ca_topic_score_gemma":0.01253491,"domain_scores_codex":[0.9985418,0.0001925705,0.00008638625,0.0002821964,0.0007338882,0.0001631578],"domain_scores_gemma":[0.9983431,0.0004444503,0.00005723277,0.0006839784,0.0003746852,0.00009658579],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0007555754,0.0001817051,0.001361492,0.0005874762,0.0002197798,0.0002714173,0.0002410615,0.01980641,0.006384705,0.03782484,0.4962749,0.4360906],"study_design_scores_gemma":[0.0005218893,0.0001371083,0.00235499,0.0002854347,0.00007288021,0.0005044235,0.0001116432,0.4891892,0.03639976,0.1667762,0.3033307,0.0003156856],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"software","genre_scores_codex":[0.003082216,0.0004908565,0.700788,0.0003210919,0.0004945722,0.0001272732,0.006766553,0.2794664,0.008463024],"genre_scores_gemma":[0.09250773,0.0008040842,0.6898301,0.0006744167,0.000262232,0.0007392038,0.03170084,0.1389218,0.04455961],"genre_candidate":"software","genre_consensus":null,"teacher_disagreement_score":0.1113579,"threshold_uncertainty_score":0.3725294,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01989926499678019,"score_gpt":0.2261883662038573,"score_spread":0.2062891012070771,"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."}}