{"id":"W4298167153","doi":"10.48550/arxiv.1905.12095","title":"On Linear Programming for Constrained and Unconstrained Average-Cost Markov Decision Processes with Countable Action Spaces and Strictly Unbounded Costs","year":2019,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Risk and Portfolio Optimization","field":"Decision Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Alberta Innovates; Alberta Innovates - Technology Futures; DeepMind; Alberta Machine Intelligence Institute","keywords":"Mathematics; Countable set; Markov decision process; Action (physics); Duality (order theory); State space; Mathematical optimization; Markov kernel; Discrete mathematics; Markov chain; Applied mathematics; Markov process; Markov model; Variable-order Markov model; Statistics","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0009591537,0.0004246593,0.0005698595,0.0005822205,0.0003774164,0.00073583,0.0004049794,0.0003654426,0.00003667234],"category_scores_gemma":[0.001208328,0.0003515779,0.00008045566,0.0009546536,0.0003393914,0.0005883082,0.0002322456,0.0003419285,0.00001165963],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001631958,"about_ca_system_score_gemma":0.0008030405,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000117446,"about_ca_topic_score_gemma":0.0006477847,"domain_scores_codex":[0.9973453,0.0001101783,0.0004012344,0.001392136,0.0003796838,0.0003714158],"domain_scores_gemma":[0.9948469,0.002703879,0.0007004334,0.0005601128,0.0009712671,0.0002173825],"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":[0.007747031,0.0004016023,0.07230164,0.0005207738,0.0003425273,0.0001436852,0.0006710465,0.7790681,0.0000313688,0.01235273,0.001386839,0.1250327],"study_design_scores_gemma":[0.01483274,0.002745092,0.005699578,0.001923503,0.0008029247,0.0001121377,0.006041558,0.8870568,0.0005081412,0.05383803,0.02354538,0.002894159],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6206557,0.00006649765,0.3758076,0.00005670058,0.0002656524,0.001840067,0.0001540522,0.00007335062,0.001080398],"genre_scores_gemma":[0.9903362,0.001055696,0.006601551,0.00005385749,0.00005549316,0.000008799229,0.0001057928,0.00003069839,0.001751899],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3696805,"threshold_uncertainty_score":0.9998936,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1095514681204798,"score_gpt":0.2775900175842334,"score_spread":0.1680385494637536,"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."}}