{"id":"W2952014622","doi":"10.48550/arxiv.1503.04269","title":"An Emphatic Approach to the Problem of Off-policy Temporal-Difference Learning","year":2015,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Temporal difference learning; Discounting; Lambda; Bootstrapping (finance); Computer science; Computation; Parametric statistics; Function (biology); Algorithm; State (computer science); Applied mathematics; Artificial intelligence; Mathematics; Reinforcement learning; Statistics; Econometrics","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":[0.0007978021,0.0003600472,0.0004273695,0.0004021668,0.0001908072,0.0002037276,0.004447334,0.0002253194,0.000003895155],"category_scores_gemma":[0.0001000799,0.0003178243,0.0001363502,0.001184109,0.000139829,0.0003183641,0.002723292,0.0008189343,0.00005234472],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002590892,"about_ca_system_score_gemma":0.0007072968,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003985924,"about_ca_topic_score_gemma":0.00001092026,"domain_scores_codex":[0.9975113,0.0004462444,0.0003416759,0.0009856123,0.0002744502,0.0004407389],"domain_scores_gemma":[0.996845,0.00009351547,0.0005517593,0.001840252,0.0003848261,0.0002846135],"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":[0.00001050208,0.00006611102,0.003668372,0.00009189475,0.0000447003,0.000008214167,0.002093564,0.9437818,0.00001267081,0.04856566,0.0001087272,0.001547802],"study_design_scores_gemma":[0.0002334278,0.0001993065,0.001031163,0.00008890172,0.00003512931,0.000002949694,0.0002588921,0.9934338,0.00002143293,0.003299136,0.001036882,0.000359011],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05425967,0.00002405268,0.935317,0.0001387608,0.0001129122,0.0006248875,0.000002448468,0.0002231478,0.009297126],"genre_scores_gemma":[0.975976,0.00003174384,0.01932789,0.00008326304,0.00008446246,0.000002943886,0.00002214636,0.0000240144,0.004447551],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9217163,"threshold_uncertainty_score":0.9999274,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09377041807705082,"score_gpt":0.2191130987965938,"score_spread":0.125342680719543,"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."}}