{"id":"W2134042548","doi":"","title":"Convergent Temporal-Difference Learning with Arbitrary Smooth Function Approximation","year":2009,"lang":"en","type":"article","venue":"Neural Information Processing Systems","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":168,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta; McGill University","funders":"","keywords":"Temporal difference learning; Function approximation; Stochastic approximation; Markov decision process; Mathematics; Function (biology); Stochastic gradient descent; Bellman equation; Mathematical optimization; Convergence (economics); Nonlinear system; Approximation algorithm; Artificial neural network; Gradient descent; Approximation error; Markov process; Rate of convergence; Algorithm; Applied mathematics; Reinforcement learning; Computer science; Artificial intelligence; Key (lock)","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":[0.002428547,0.0008440595,0.001028083,0.0005100523,0.0004018143,0.001091728,0.001799633,0.001451297,0.002872385],"category_scores_gemma":[0.008947497,0.0004719117,0.0008248786,0.0005218561,0.00176298,0.001887123,0.002068795,0.002300656,0.0005582396],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001347799,"about_ca_system_score_gemma":0.001307006,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003179274,"about_ca_topic_score_gemma":0.001936952,"domain_scores_codex":[0.9993286,0.0001756819,0.00004290179,0.0001358357,0.0002541276,0.00006286525],"domain_scores_gemma":[0.9976478,0.001377467,0.0001821658,0.000255222,0.0004383454,0.00009886328],"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":[0.0001182077,0.00006276508,0.0005913373,0.0001536661,0.00004692168,0.00009520556,0.0001263292,0.7048174,0.002621078,0.2241525,0.001096876,0.06611783],"study_design_scores_gemma":[0.000006469439,0.00001728161,0.00002152804,0.000005862671,0.000002701945,0.0000128072,0.00000210343,0.9857782,0.0004800543,0.01313381,0.0005346536,0.000004486166],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.003118963,0.0001842242,0.9950233,0.0001141731,0.00004114212,0.00002142456,0.00001278687,0.00008456127,0.001399549],"genre_scores_gemma":[0.3630849,0.0004497861,0.6290939,0.0002417156,0.00008954435,0.0002337302,0.0000930766,0.0001302213,0.006583024],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003179274,"threshold_uncertainty_score":0.01284355,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01454759173183138,"score_gpt":0.2203356894460717,"score_spread":0.2057880977142404,"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."}}