{"id":"W2064203730","doi":"10.5555/777092.777139","title":"Greedy linear value-approximation for factored Markov decision processes","year":2002,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":41,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; University of Waterloo","funders":"","keywords":"Markov decision process; Bellman equation; Mathematical optimization; Linear programming; Computer science; Dynamic programming; Basis (linear algebra); Set (abstract data type); Linear approximation; Approximation error; Approximation algorithm; Function (biology); Value (mathematics); Function approximation; Markov process; Mathematics; Artificial intelligence; Nonlinear system; Artificial neural network","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":[],"consensus_categories":[],"category_scores_codex":[0.000192358,0.0001258902,0.0001200661,0.00009667777,0.0001346835,0.0001781694,0.0006357001,0.00006442634,0.00009152351],"category_scores_gemma":[0.0005385348,0.0001034151,0.00004789832,0.0003608294,0.00001879308,0.0006798207,0.0001242405,0.00006974959,0.0001344098],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003494412,"about_ca_system_score_gemma":0.00002321102,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000003864592,"about_ca_topic_score_gemma":0.000001089805,"domain_scores_codex":[0.998827,0.00001698882,0.0002613247,0.0002932274,0.000357463,0.0002440361],"domain_scores_gemma":[0.9988984,0.0003207932,0.0001067288,0.0004006804,0.0002088762,0.00006451199],"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":[0.00003139403,0.0001927873,0.001474396,0.0003990686,0.00005124836,0.000003897678,0.001995924,0.7220168,0.0004747547,0.04489735,0.02384994,0.2046125],"study_design_scores_gemma":[0.0003544364,0.0001340586,0.00017845,0.00002254657,0.000004062566,0.000002826167,0.00001146982,0.989593,0.00132173,0.0008697044,0.007363641,0.0001441307],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.001101432,0.00003501176,0.9926317,0.000355246,0.0002299873,0.0003870401,6.44676e-7,0.0002759452,0.004983028],"genre_scores_gemma":[0.203998,0.00002586096,0.7908021,0.0002441861,0.00007595152,0.00003152013,0.000006102603,0.00001365217,0.004802618],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.2675762,"threshold_uncertainty_score":0.4217144,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03629519433379755,"score_gpt":0.266278795498935,"score_spread":0.2299836011651375,"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."}}