{"id":"W3035490956","doi":"10.48550/arxiv.2006.06903","title":"On Correctness of Automatic Differentiation for Non-Differentiable Functions","year":2020,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Model Reduction and Neural Networks","field":"Physics and Astronomy","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"","keywords":"Differentiable function; Correctness; Automatic differentiation; Counterexample; Computer science; Set (abstract data type); Artificial intelligence; Function (biology); Artificial neural network; Point (geometry); Theoretical computer science; Algebra over a field; Mathematics; Pure mathematics; Algorithm; Discrete mathematics; Programming language","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":[0.00003204209,0.000218555,0.0003282368,0.0001000279,0.0001094586,0.00002463258,0.0002149046,0.0001059812,0.0005238606],"category_scores_gemma":[0.000004388375,0.0002365228,0.0002999538,0.0001700156,0.00003708165,0.00005789676,0.0001451921,0.0002911213,0.00002467689],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003780479,"about_ca_system_score_gemma":0.00005785746,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004708692,"about_ca_topic_score_gemma":0.000002335084,"domain_scores_codex":[0.9990836,0.00003874076,0.0001849899,0.000471872,0.00005080005,0.0001700483],"domain_scores_gemma":[0.9991225,0.00008977758,0.0002793916,0.0003227432,0.00009272007,0.00009284569],"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":[0.0003374398,0.0008198837,0.005667015,0.0005695611,0.0007900917,0.000002511081,0.000222665,0.7967574,0.0006408995,0.175242,0.01428281,0.004667734],"study_design_scores_gemma":[0.00064769,0.00007783742,0.0008103458,0.0001368672,0.0002398684,6.646145e-8,0.000105131,0.9600518,0.0007956962,0.03683254,0.00006700996,0.0002351591],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5385582,0.000001649658,0.4591519,0.0000278869,0.0006024652,0.0003411353,0.00008705709,0.00003876075,0.001190931],"genre_scores_gemma":[0.9966625,0.00000389154,0.00005479301,0.00001464739,0.0001880378,0.000007297692,0.0003483551,0.00002208088,0.002698378],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4590971,"threshold_uncertainty_score":0.9645118,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06257417997478137,"score_gpt":0.1943270186608312,"score_spread":0.1317528386860498,"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."}}