Mots-clés
Markov decision processes; gambling houses; POMDPs; repeated games; distance for belief spaces; Kantorovich-Rubinstein duality; Wasserstein metric; limit value; uniform value; general values; characterization of the value;
Remplace
Jérôme Renault et Xavier Venel, « A distance for probability spaces, and long-term values in Markov Decision Processes and Repeated Games », TSE Working Paper, n° 17-748, janvier 2017.
Référence
Jérôme Renault et Xavier Venel, « A distance for probability spaces, and long-term values in Markov Decision Processes and Repeated Games », Mathematics of Operations Research, vol. 42, n° 2, 2017, p. 349–376.
Voir aussi
Publié dans
Mathematics of Operations Research, vol. 42, n° 2, 2017, p. 349–376