{"id":"W6979372690","doi":"","title":"PERRY: Policy Evaluation with Confidence Intervals using Auxiliary Data","year":2025,"lang":"en","type":"article","venue":"ArXiv.org","topic":"Injection Molding Process and Properties","field":"Engineering","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Institutes of Health; National Human Genome Research Institute; Parker Institute for Cancer Immunotherapy; Canadian Institute for Advanced Research","keywords":"Task (project management); Confidence interval; Reinforcement learning; Value (mathematics); Ground truth; Construct (python library)","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":[],"consensus_categories":[],"category_scores_codex":[0.0003088604,0.0001080033,0.0001102577,0.0001198693,0.0001007868,0.00004858356,0.0002416337,0.00004721981,0.00008902374],"category_scores_gemma":[0.0001046636,0.00009040908,0.00001446847,0.0002892682,0.0000381003,0.0003495656,0.00008196745,0.0001191345,0.00003016373],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008914711,"about_ca_system_score_gemma":0.0001532286,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003103236,"about_ca_topic_score_gemma":0.00003948532,"domain_scores_codex":[0.9993424,0.00002943983,0.0001470654,0.0001915004,0.0001360948,0.0001534998],"domain_scores_gemma":[0.999431,0.00002101505,0.00002326824,0.0003923536,0.0001081119,0.00002425994],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001987523,0.00009804386,0.5140435,0.001716734,0.0007487672,0.00001489699,0.00462015,0.3396017,0.06282509,0.003374818,0.007666753,0.06509073],"study_design_scores_gemma":[0.0007770363,0.00007936868,0.07051565,0.0009720074,0.0002018479,0.00002544373,0.0007818565,0.8832344,0.02871172,0.0005210605,0.01364785,0.0005316977],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9841733,0.0009643287,0.01023256,0.0002563227,0.0003150624,0.0001431699,0.000006789021,0.0001827483,0.003725717],"genre_scores_gemma":[0.9988358,0.00005527701,0.0003200732,0.0001993787,0.0001130183,0.00001215175,0.0000142821,0.0000167355,0.0004332861],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5436327,"threshold_uncertainty_score":0.3686775,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.141702541937224,"score_gpt":0.3492679357619882,"score_spread":0.2075653938247643,"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."}}