{"id":"W4403580911","doi":"10.48550/arxiv.2410.13577","title":"Generalization Bounds via Meta-Learned Model Representations: PAC-Bayes and Sample Compression Hypernetworks","year":2024,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Machine Learning and Data Classification","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Fonds de recherche du Québec – Nature et technologies; Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Generalization; Sample (material); Compression (physics); Artificial intelligence; Computer science; Mathematics; Physics; 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":[0.0003399838,0.000303174,0.0003482328,0.0002764773,0.0002712545,0.0004758726,0.0008400736,0.0002657805,0.00002682208],"category_scores_gemma":[0.00009074001,0.0003008011,0.0001705936,0.000502917,0.00008905648,0.0003846524,0.002238853,0.0006429208,0.0000281632],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008147372,"about_ca_system_score_gemma":0.0001056234,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006160666,"about_ca_topic_score_gemma":0.00006662925,"domain_scores_codex":[0.9977326,0.0002829652,0.0002465606,0.001364244,0.0001358348,0.000237803],"domain_scores_gemma":[0.9980239,0.0002179082,0.0002453151,0.001247118,0.0001267572,0.0001390431],"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":[0.000008335491,0.00003413042,0.0008516784,0.00007884279,0.0002267399,0.00001190898,0.0002202683,0.8098552,0.0000543911,0.1855313,0.0009251619,0.002202062],"study_design_scores_gemma":[0.0001267729,0.00001286594,0.0003159458,0.00003305874,0.0003574219,0.000002861834,0.00001422181,0.7718757,0.00001661054,0.2264572,0.0005492055,0.0002380727],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01688439,0.000734929,0.9799543,0.0006130348,0.0002899215,0.0002314239,0.00004574568,0.0004366048,0.0008096466],"genre_scores_gemma":[0.9732374,0.0008247357,0.02347442,0.0001099855,0.00008066318,0.000005214169,0.0003680589,0.00002537491,0.001874141],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9564799,"threshold_uncertainty_score":0.9999444,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1479020429098674,"score_gpt":0.2465707154410324,"score_spread":0.09866867253116499,"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."}}