{"_id":"1873","oa_version":"Preprint","date_updated":"2025-09-23T09:52:31Z","user_id":"317138e5-6ab7-11ef-aa6d-ffef3953e345","publisher":"Elsevier","isi":1,"publication_status":"published","article_processing_charge":"No","quality_controlled":"1","main_file_link":[{"url":"https://arxiv.org/abs/1408.2058","open_access":"1"}],"publist_id":"5224","month":"04","department":[{"_id":"KrCh"}],"type":"journal_article","scopus_import":"1","volume":221,"day":"01","doi":"10.1016/j.artint.2014.12.009","year":"2015","corr_author":"1","status":"public","date_created":"2018-12-11T11:54:28Z","citation":{"chicago":"Chatterjee, Krishnendu, and Martin Chmelik. “POMDPs under Probabilistic Semantics.” <i>Artificial Intelligence</i>. Elsevier, 2015. <a href=\"https://doi.org/10.1016/j.artint.2014.12.009\">https://doi.org/10.1016/j.artint.2014.12.009</a>.","mla":"Chatterjee, Krishnendu, and Martin Chmelik. “POMDPs under Probabilistic Semantics.” <i>Artificial Intelligence</i>, vol. 221, Elsevier, 2015, pp. 46–72, doi:<a href=\"https://doi.org/10.1016/j.artint.2014.12.009\">10.1016/j.artint.2014.12.009</a>.","apa":"Chatterjee, K., &#38; Chmelik, M. (2015). POMDPs under probabilistic semantics. <i>Artificial Intelligence</i>. Elsevier. <a href=\"https://doi.org/10.1016/j.artint.2014.12.009\">https://doi.org/10.1016/j.artint.2014.12.009</a>","short":"K. Chatterjee, M. Chmelik, Artificial Intelligence 221 (2015) 46–72.","ista":"Chatterjee K, Chmelik M. 2015. POMDPs under probabilistic semantics. Artificial Intelligence. 221, 46–72.","ama":"Chatterjee K, Chmelik M. POMDPs under probabilistic semantics. <i>Artificial Intelligence</i>. 2015;221:46-72. doi:<a href=\"https://doi.org/10.1016/j.artint.2014.12.009\">10.1016/j.artint.2014.12.009</a>","ieee":"K. Chatterjee and M. Chmelik, “POMDPs under probabilistic semantics,” <i>Artificial Intelligence</i>, vol. 221. Elsevier, pp. 46–72, 2015."},"intvolume":"       221","oa":1,"date_published":"2015-04-01T00:00:00Z","author":[{"last_name":"Chatterjee","orcid":"0000-0002-4561-241X","id":"2E5DCA20-F248-11E8-B48F-1D18A9856A87","first_name":"Krishnendu","full_name":"Chatterjee, Krishnendu"},{"id":"3624234E-F248-11E8-B48F-1D18A9856A87","last_name":"Chmelik","full_name":"Chmelik, Martin","first_name":"Martin"}],"abstract":[{"text":"We consider partially observable Markov decision processes (POMDPs) with limit-average payoff, where a reward value in the interval [0,1] is associated with every transition, and the payoff of an infinite path is the long-run average of the rewards. We consider two types of path constraints: (i) a quantitative constraint defines the set of paths where the payoff is at least a given threshold λ1ε(0,1]; and (ii) a qualitative constraint which is a special case of the quantitative constraint with λ1=1. We consider the computation of the almost-sure winning set, where the controller needs to ensure that the path constraint is satisfied with probability 1. Our main results for qualitative path constraints are as follows: (i) the problem of deciding the existence of a finite-memory controller is EXPTIME-complete; and (ii) the problem of deciding the existence of an infinite-memory controller is undecidable. For quantitative path constraints we show that the problem of deciding the existence of a finite-memory controller is undecidable. We also present a prototype implementation of our EXPTIME algorithm and experimental results on several examples.","lang":"eng"}],"page":"46 - 72","external_id":{"isi":["000350782300003"],"arxiv":["1408.2058"]},"title":"POMDPs under probabilistic semantics","publication":"Artificial Intelligence","language":[{"iso":"eng"}],"arxiv":1}