{"id":"W7043660303","doi":"","title":"Stochastic variance reduced gradient for policy evaluation with fewer gradient evaluations","year":2020,"lang":"en","type":"dissertation","venue":"eScholarship@McGill (McGill)","topic":"Stochastic Gradient Optimization Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Reinforcement learning; Convergence (economics); Gradient method; Computation; Variance (accounting); Function (biology); Gradient descent; Bellman equation","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00268377,0.001338538,0.001719314,0.0008685916,0.000590653,0.00151728,0.001340362,0.001840833,0.00709319],"category_scores_gemma":[0.01104226,0.0009152709,0.001146571,0.000850986,0.0009527926,0.00167647,0.001761339,0.003077757,0.002623298],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001194394,"about_ca_system_score_gemma":0.002446214,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004707941,"about_ca_topic_score_gemma":0.005696428,"domain_scores_codex":[0.9981921,0.000730937,0.0001031289,0.0003036854,0.0005221754,0.0001478818],"domain_scores_gemma":[0.9970725,0.001627063,0.000172887,0.0004791386,0.0005436699,0.0001047361],"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":[0.0002037009,0.0001449729,0.0006431467,0.0001832017,0.00009539067,0.0001261586,0.00009207947,0.7343283,0.005562909,0.06142438,0.01268843,0.1845073],"study_design_scores_gemma":[0.00001047023,0.00001585035,0.00005249573,0.0000086381,0.000003755916,0.00001300001,0.000002670213,0.9917607,0.0006934832,0.00624474,0.001188964,0.000005229663],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.002674319,0.0001491153,0.9942445,0.000183888,0.00007806942,0.00004868957,0.00004150243,0.0008769905,0.001702937],"genre_scores_gemma":[0.1453021,0.0002128395,0.845787,0.0003617444,0.0001313134,0.0004029791,0.0003939446,0.0008813213,0.00652665],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.00709319,"threshold_uncertainty_score":0.02372909,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03332380938638713,"score_gpt":0.3014796231985526,"score_spread":0.2681558138121655,"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."}}