{"id":"W3093671052","doi":"10.48550/arxiv.2010.11924","title":"In Search of Robust Measures of Generalization","year":2020,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; Université de Montréal; Université Laval","funders":"","keywords":"Generalization; Robustness (evolution); Computer science; Generalization error; VC dimension; Artificial intelligence; Artificial neural network; Machine learning; Econometrics; Mathematics","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":[0.02739443,0.002235272,0.002505849,0.004974163,0.001147174,0.003752644,0.004037938,0.003951883,0.002161155],"category_scores_gemma":[0.1442965,0.001021939,0.002053889,0.002944425,0.008692928,0.01315667,0.006912412,0.00684288,0.0005896897],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002983928,"about_ca_system_score_gemma":0.001537317,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001696987,"about_ca_topic_score_gemma":0.001003637,"domain_scores_codex":[0.9860248,0.005339613,0.001011122,0.003744873,0.003342861,0.0005368175],"domain_scores_gemma":[0.8942927,0.06981902,0.009522809,0.02126768,0.003813025,0.001284802],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.000232166,0.0001227353,0.01230208,0.0008039619,0.001006555,0.0001651589,0.0006000922,0.3896791,0.002305363,0.5019147,0.003818063,0.08705001],"study_design_scores_gemma":[0.00001572121,0.0001401366,0.002948862,0.000147338,0.00005811534,0.0001003861,0.00007843608,0.3271229,0.001222169,0.6660938,0.002015135,0.00005682739],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.03743448,0.003941188,0.9488292,0.002633575,0.0001612429,0.0001026964,0.000497877,0.000674425,0.005725324],"genre_scores_gemma":[0.8151488,0.003078545,0.1738164,0.001929067,0.0008078003,0.000647658,0.001436289,0.0006325172,0.002502849],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.02739443,"threshold_uncertainty_score":0.1448773,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2157159118589783,"score_gpt":0.216649673032837,"score_spread":0.0009337611738587381,"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."}}