{"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":"codex-gemma-dda1882f352a","candidate_categories":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0002162809,0.0000766079,0.00007980729,0.0001384229,0.0005265065,0.001212815,0.0004084179,0.00003108102,0.0003347436],"category_scores_gemma":[0.00007774661,0.00007844831,0.00001316259,0.0008083069,0.00004794495,0.0008316778,0.001064377,0.0001355541,0.0002141653],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003053764,"about_ca_system_score_gemma":0.000002328227,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001168058,"about_ca_topic_score_gemma":7.554083e-7,"domain_scores_codex":[0.9991353,0.00009918311,0.0001625068,0.0002357738,0.0001772711,0.0001900156],"domain_scores_gemma":[0.9994094,0.00002065735,0.00005348038,0.0002429768,0.0001720114,0.0001014422],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0000114447,0.00008156802,0.0001737523,0.00006364605,0.00001387047,0.00002043879,0.001362346,0.0002380051,0.00140037,0.02298409,0.03484928,0.9388012],"study_design_scores_gemma":[0.0004639268,0.00008069367,0.02041312,0.0000227949,0.000004399546,0.0001845918,0.0001790806,0.03526686,0.000265804,0.0004830558,0.9424434,0.0001922799],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7414504,0.002170866,0.153387,0.01763089,0.0003828144,0.001563644,0.0002292904,0.001786965,0.08139811],"genre_scores_gemma":[0.9975642,0.0003622632,0.001086948,0.0003690948,0.00005386854,9.252222e-8,0.0001926496,0.0001244471,0.0002464404],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9386089,"threshold_uncertainty_score":0.999824,"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."}}