{"id":"W4406610289","doi":"10.1109/tac.2025.3532181","title":"Model Approximation in MDPs With Unbounded Per-Step Cost","year":2025,"lang":"en","type":"article","venue":"IEEE Transactions on Automatic Control","topic":"Advanced Bandit Algorithms Research","field":"Decision Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"Natural Sciences and Engineering Research Council of Canada; Innovation for Defence Excellence and Security","keywords":"Computer science; Mathematical optimization; Mathematics; Applied mathematics","routes":{"ca_aff":true,"ca_fund":true,"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":[0.001371138,0.0002323916,0.0004788059,0.001129291,0.0002684567,0.0002668196,0.0005707187,0.0001155592,0.0002701118],"category_scores_gemma":[0.0001897131,0.0001713959,0.0001153863,0.001624169,0.0001649388,0.0003942808,0.000002295991,0.0003469005,0.0002261732],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003275093,"about_ca_system_score_gemma":0.0003943356,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004472319,"about_ca_topic_score_gemma":0.0009539617,"domain_scores_codex":[0.9967366,0.0002670648,0.0007587632,0.0005667132,0.001238704,0.0004321894],"domain_scores_gemma":[0.9968254,0.00185526,0.0001377818,0.0007647774,0.0003089305,0.0001078633],"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":[0.0001695648,0.0002674978,0.00002441062,0.00001623823,0.00003793978,0.000006063645,0.0002203817,0.5958198,0.0002734851,0.0002044613,0.00007925335,0.4028809],"study_design_scores_gemma":[0.003840285,0.0001067126,0.0003416823,0.00007543289,0.00002548454,0.000005200708,0.0003059525,0.9858029,0.0006674491,0.008536443,0.0001265596,0.0001659115],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01110723,0.00001738146,0.9847388,0.001218046,0.0001889798,0.001459824,0.00003788275,0.0001530505,0.001078785],"genre_scores_gemma":[0.9771931,0.000004504891,0.01861213,0.0002585753,0.000009277078,0.0005756628,0.000001254365,0.00001819886,0.003327258],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9661267,"threshold_uncertainty_score":0.6989323,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05376352359343873,"score_gpt":0.3732842671564823,"score_spread":0.3195207435630436,"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."}}