{"id":"W4319452827","doi":"10.48550/arxiv.2302.02432","title":"Tighter Information-Theoretic Generalization Bounds from Supersamples","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Image Retrieval and Classification Techniques","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Generalization; Mathematical economics; Mathematics; Computer science; Econometrics; Mathematical analysis","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0001732849,0.0002292724,0.0002041184,0.0003050836,0.0001620117,0.0004082003,0.001457948,0.0002650033,0.00007515673],"category_scores_gemma":[0.0000481583,0.0002483187,0.0001439015,0.0006169909,0.0001147142,0.001070545,0.001051942,0.000271132,0.0004751412],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001656956,"about_ca_system_score_gemma":0.0001606848,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003560521,"about_ca_topic_score_gemma":0.00001517343,"domain_scores_codex":[0.998752,0.0001010033,0.0002453639,0.0005502526,0.0001276125,0.0002237449],"domain_scores_gemma":[0.9983494,0.00008764278,0.0002126569,0.001006954,0.0002472967,0.00009601493],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001027347,0.00002855313,0.001171365,0.00005661636,0.00006236842,0.00002259988,0.0006588249,0.002759465,0.00008904032,0.9915084,0.002194573,0.00143793],"study_design_scores_gemma":[0.0002180089,0.00002381634,0.002463105,0.00007050893,0.0000481492,9.337273e-7,0.00007040941,0.574809,0.002584107,0.4120813,0.007119998,0.000510689],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01771539,0.00002348803,0.9791017,0.0003608094,0.0005014273,0.0002095646,0.00007001738,0.00103325,0.0009843807],"genre_scores_gemma":[0.9897773,0.0004449174,0.007407859,0.0002311673,0.00008373865,0.000003120996,0.0004512916,0.00001634668,0.001584283],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9720619,"threshold_uncertainty_score":0.9999969,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07705969476319298,"score_gpt":0.1916398359882436,"score_spread":0.1145801412250507,"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."}}