{"id":"W4387559563","doi":"10.48550/arxiv.2310.04935","title":"Statistical Guarantees for Variational Autoencoders using PAC-Bayesian Theory","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Generative Adversarial Networks and Image Synthesis","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université Laval","funders":"Institut de Valorisation des Données; Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Generalization; Bayesian probability; Computer science; Generative grammar; Upper and lower bounds; Posterior probability; Distribution (mathematics); Artificial intelligence; Generative model; Statistical learning theory; Algorithm; Machine learning; Mathematics","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0005903231,0.0003134933,0.0003508938,0.0002170087,0.0003251769,0.0002062532,0.001113672,0.0002326468,0.00004649414],"category_scores_gemma":[0.000186835,0.0003480357,0.0002301384,0.0003577586,0.000141933,0.0003255771,0.0008865293,0.0002942198,0.00002166004],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001840784,"about_ca_system_score_gemma":0.0003949096,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00009399339,"about_ca_topic_score_gemma":0.00003261586,"domain_scores_codex":[0.9978194,0.0003050182,0.0002367376,0.001091199,0.0001177517,0.0004299163],"domain_scores_gemma":[0.9978951,0.0008744521,0.0002144728,0.000686648,0.000198886,0.0001304537],"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":[0.00002989017,0.00002703207,0.00006189085,0.00002323874,0.0001192609,0.00004673475,0.0001244867,0.6055384,0.00001614017,0.3931273,0.000603522,0.0002820925],"study_design_scores_gemma":[0.0002360757,0.00002532662,0.0002616875,0.00003217743,0.00007768413,0.000001222605,0.00005014239,0.7055178,0.00002293264,0.2933437,0.000158171,0.0002730215],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0007218533,0.00002097006,0.9968599,0.0001894466,0.001195519,0.0003887195,0.0001722933,0.0002305127,0.0002207589],"genre_scores_gemma":[0.7828049,0.00002162723,0.2158769,0.0001142105,0.0002554158,0.000001834779,0.00005534204,0.00003025051,0.0008394972],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.782083,"threshold_uncertainty_score":0.9998972,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1127911845074769,"score_gpt":0.2227502722686008,"score_spread":0.1099590877611239,"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."}}