{"id":"W3014636579","doi":"10.48550/arxiv.2004.00353","title":"SUMO: Unbiased Estimation of Log Marginal Probability for Latent\\n Variable Models","year":2020,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Latent variable; Estimator; Mathematics; Latent variable model; Statistics; Parameterized complexity; Truncation (statistics); Minimum-variance unbiased estimator; Bias of an estimator; Efficient estimator; Variable (mathematics); Algorithm","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":[0.0005214624,0.0003110212,0.0004928897,0.0001593948,0.0001355941,0.00007251514,0.001798456,0.0002894493,0.00001608424],"category_scores_gemma":[0.0002565061,0.0003761827,0.0002197107,0.0005905917,0.0001154479,0.0006144616,0.001815471,0.0005707697,0.000007595535],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002439441,"about_ca_system_score_gemma":0.0004165812,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001627447,"about_ca_topic_score_gemma":0.000005334564,"domain_scores_codex":[0.997816,0.000199525,0.0003284464,0.001189009,0.0001379627,0.0003290843],"domain_scores_gemma":[0.9977218,0.0002929505,0.0005169789,0.001041518,0.0002890699,0.0001377175],"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":[0.00006236919,0.00004646639,0.0003093279,0.0002971008,0.00004099136,0.00001104314,0.00008975586,0.6291395,0.0000143882,0.3697432,0.00003815863,0.0002076498],"study_design_scores_gemma":[0.0004025255,0.00004414072,0.00009853506,0.0000638485,0.00005754552,6.407833e-7,0.000005773624,0.5982738,0.00004147577,0.4008005,0.00001536542,0.0001958206],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01813765,0.00001197477,0.9794894,0.0002237525,0.0003907772,0.0008029334,0.00003548027,0.0002468812,0.0006611824],"genre_scores_gemma":[0.7611458,0.000005704339,0.2385771,0.0000286926,0.00003762391,0.00000301333,0.00004687465,0.00001688787,0.000138289],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7430081,"threshold_uncertainty_score":0.999869,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1211155253638014,"score_gpt":0.2156941618349301,"score_spread":0.09457863647112869,"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."}}