{"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":"codex-gemma-dda1882f352a","candidate_categories":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0003086166,0.0002223608,0.0002009777,0.0002202125,0.0003804546,0.001254692,0.000394027,0.00009119268,0.000002955897],"category_scores_gemma":[0.00003746253,0.0001781739,0.0000342085,0.0005714304,0.00002908029,0.005834027,0.00003667051,0.0003370327,0.00005803531],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008591708,"about_ca_system_score_gemma":0.0001059508,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001664588,"about_ca_topic_score_gemma":2.602174e-7,"domain_scores_codex":[0.9981189,0.00007744353,0.0006156633,0.0002124372,0.0006704886,0.0003050246],"domain_scores_gemma":[0.9985921,0.00002447851,0.0007143242,0.0002707612,0.0003032678,0.00009501889],"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":[0.00003811504,0.00002140727,0.002064663,0.0003278746,0.00001071911,0.000001978687,0.002588566,0.8960687,0.0002584095,0.007980965,0.000198037,0.09044057],"study_design_scores_gemma":[0.000384239,0.0004389012,0.004044361,0.0001566986,0.000007907748,0.0000324485,0.0002760931,0.9909536,0.0001483234,0.00005426573,0.003259605,0.0002435369],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01403097,0.000066004,0.9802634,0.0002932832,0.0003826908,0.0003897695,2.952454e-7,0.0007391103,0.003834518],"genre_scores_gemma":[0.9948147,0.000003605885,0.004097127,0.0004879033,0.0000631717,0.00002462385,0.00005705701,0.000007337842,0.0004444992],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9807837,"threshold_uncertainty_score":0.9997821,"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."}}