{"id":"W3167624337","doi":"","title":"Sparse Feature Selection Makes Batch Reinforcement Learning More Sample Efficient","year":2021,"lang":"en","type":"article","venue":"International Conference on Machine Learning","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Minimax; Estimator; Reinforcement learning; Lasso (programming language); Upper and lower bounds; Covariance; Divergence (linguistics); Computer science; Sample size determination; Feature selection; Mathematical optimization; Dimension (graph theory); Mathematics; Function (biology); Model selection; Algorithm; Artificial intelligence; Statistics","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":[0.004051093,0.0009791417,0.001640496,0.0003790897,0.0004839152,0.001129911,0.001512322,0.001292501,0.002394963],"category_scores_gemma":[0.01881004,0.0006761109,0.0006048782,0.0003977407,0.001954983,0.002139019,0.001858807,0.002776736,0.0005918536],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001195826,"about_ca_system_score_gemma":0.001957601,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002682559,"about_ca_topic_score_gemma":0.00231581,"domain_scores_codex":[0.9981766,0.000674391,0.00008636634,0.0003895245,0.000513034,0.000160088],"domain_scores_gemma":[0.9912424,0.006194307,0.0006494965,0.0009898276,0.0006859619,0.0002379808],"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":[0.0002174231,0.0001524112,0.0009786143,0.00008564788,0.00004936723,0.00006185281,0.0000725378,0.904862,0.004405987,0.03318214,0.001656164,0.05427589],"study_design_scores_gemma":[0.00001145059,0.00003016561,0.00008144597,0.000004652321,0.00000393378,0.000008192198,0.000003467366,0.9912151,0.0006456783,0.007745363,0.0002463606,0.000004180034],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.007920052,0.00007753014,0.9906484,0.0001844723,0.00002200988,0.00003686429,0.00001949047,0.0003108042,0.0007802336],"genre_scores_gemma":[0.661034,0.0001567147,0.3347886,0.0004005359,0.0001043732,0.0003266913,0.0001704262,0.0003149529,0.002703643],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004051093,"threshold_uncertainty_score":0.02142447,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03003771541059881,"score_gpt":0.2933460117824498,"score_spread":0.263308296371851,"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."}}