{"id":"W4382141796","doi":"10.48550/arxiv.2306.13284","title":"Correcting discount-factor mismatch in on-policy policy gradient methods","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Age of Information Optimization","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Estimator; Discounting; Distribution (mathematics); Variance (accounting); Applied mathematics; Mathematical optimization; Mathematics; Computer science; Statistics; Mathematical analysis; Economics; Finance","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005660394,0.002057824,0.002422948,0.0009515248,0.0007037323,0.001996571,0.003169861,0.00237396,0.003458638],"category_scores_gemma":[0.03520822,0.001021322,0.0007786864,0.000843881,0.00156603,0.003057048,0.002965431,0.004439165,0.001238526],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001885866,"about_ca_system_score_gemma":0.003983761,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006032485,"about_ca_topic_score_gemma":0.005934082,"domain_scores_codex":[0.9965938,0.00120356,0.0002394258,0.0006892246,0.000942124,0.000331967],"domain_scores_gemma":[0.9924385,0.004340124,0.0005419203,0.001165155,0.001260809,0.0002535032],"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":[0.000590297,0.0002638183,0.002746475,0.0002855679,0.0001781935,0.0001481776,0.0002037413,0.5986835,0.00386878,0.03129083,0.0084329,0.3533077],"study_design_scores_gemma":[0.00004107533,0.00005476626,0.0001752781,0.00002627821,0.00001755113,0.00003522371,0.00001127733,0.9833028,0.002806552,0.01193325,0.001581006,0.00001495649],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01544643,0.0008182333,0.9782299,0.0004047946,0.0003232488,0.0001114143,0.00006752866,0.00277534,0.001823125],"genre_scores_gemma":[0.6209238,0.000687007,0.3684347,0.0009103377,0.0003514821,0.0002750568,0.0004468954,0.001400274,0.006570548],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006032485,"threshold_uncertainty_score":0.02993542,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1144003186648897,"score_gpt":0.2718175409360546,"score_spread":0.1574172222711649,"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."}}