{"id":"W3182076562","doi":"10.1111/poms.13778","title":"Sublinear regret for learning POMDPs","year":2022,"lang":"en","type":"article","venue":"Production and Operations Management","topic":"Advanced Bandit Algorithms Research","field":"Decision Sciences","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Regret; Markov decision process; Reinforcement learning; Oracle; Computer science; Sublinear function; Partially observable Markov decision process; Mathematical optimization; Upper and lower bounds; Time horizon; Artificial intelligence; Markov chain; Machine learning; Markov process; Markov model; Mathematics; Statistics; Discrete 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":["sts"],"consensus_categories":[],"category_scores_codex":[0.002294044,0.00006570997,0.00008714908,0.0003361487,0.001603246,0.0002559104,0.000205844,0.000009100879,0.0004402751],"category_scores_gemma":[0.0008411942,0.00005683467,0.00003062372,0.0006661533,0.00003272689,0.00030202,0.0002982441,0.0001364701,0.00003326354],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006409312,"about_ca_system_score_gemma":0.00001742034,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000004440992,"about_ca_topic_score_gemma":0.00001270813,"domain_scores_codex":[0.9982799,0.0001449012,0.0002388493,0.0004658987,0.0007153575,0.000155068],"domain_scores_gemma":[0.999382,0.00005964645,0.00003120086,0.0002750667,0.0002092895,0.00004283994],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00006963938,0.0002040063,0.0003949247,0.00001775088,0.00004823289,0.000003840576,0.001093757,0.7380575,0.0001078093,0.01881172,0.03471005,0.2064808],"study_design_scores_gemma":[0.0002848924,0.0001407692,0.0005008031,0.00000144733,0.000008947546,0.000009455171,0.004201521,0.1394288,0.0001529533,0.002972854,0.8521903,0.0001072727],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05658526,0.0008073216,0.8675482,0.0603576,0.003381614,0.005138849,0.00003829656,0.000444455,0.005698371],"genre_scores_gemma":[0.6779448,0.0002330687,0.0721116,0.0003099435,0.0004737701,0.001628813,0.00005887519,0.00002723494,0.2472119],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8174803,"threshold_uncertainty_score":0.9996966,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1261665713380833,"score_gpt":0.4307963280159451,"score_spread":0.3046297566778619,"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."}}