{"id":"W2982044490","doi":"","title":"Detecting Extrapolation with Local Ensembles","year":2019,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Extrapolation; Spurious relationship; Hessian matrix; Computer science; Curvature; Focus (optics); Component (thermodynamics); Artificial intelligence; Algorithm; Mathematics; Pattern recognition (psychology); Machine learning; Applied mathematics; Statistics; Physics","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":[0.000288845,0.000331147,0.0003124014,0.0002654313,0.0002074257,0.000145969,0.001486386,0.0002956767,0.00001783045],"category_scores_gemma":[0.00004271532,0.0003527456,0.0001211854,0.0005293493,0.0001004886,0.0005550935,0.001661035,0.00102605,0.00007793333],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002609715,"about_ca_system_score_gemma":0.0002380084,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000202311,"about_ca_topic_score_gemma":0.00005928783,"domain_scores_codex":[0.9979774,0.0001954215,0.0001685833,0.001149385,0.0001401726,0.0003689807],"domain_scores_gemma":[0.9980451,0.0002102089,0.0003449547,0.001154098,0.0001394212,0.0001061613],"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":[0.0000319335,0.00001597336,0.005305666,0.00005594681,0.00004721296,0.0001432186,0.0002252132,0.9528749,0.00002044207,0.03814429,0.000009941356,0.003125281],"study_design_scores_gemma":[0.0004307304,0.00006332991,0.001242829,0.0001477247,0.00004575493,0.00001191889,0.0001330113,0.9901457,0.000114046,0.007094318,0.0001184708,0.0004520967],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1425454,0.0000286576,0.85417,0.00004680617,0.0004730112,0.0002200611,0.000001287792,0.0003420428,0.002172651],"genre_scores_gemma":[0.9858441,0.00001374499,0.01340266,0.00003892355,0.00008768548,4.829358e-7,0.000006328459,0.00002797636,0.0005780461],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8432987,"threshold_uncertainty_score":0.9998925,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04736355427099913,"score_gpt":0.1854851393122977,"score_spread":0.1381215850412986,"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."}}