Limits of Multi-Discounted Markov Decision Processes

Hugo Gimbert, Wieslaw Zielonka
2007 22nd Annual IEEE Symposium on Logic in Computer Science (LICS 2007)  
Markov decision processes (MDPs) are controllable discrete event systems with stochastic transitions. The payoff received by the controller can be evaluated in different ways, depending on the payoff function the MDP is equipped with. For example a mean-payoff function evaluates average performance, whereas a discounted payoff function gives more weights to earlier performance by means of a discount factor. Another well-known example is the parity payoff function which is used to encode logical
more » ... d to encode logical specifications [14] . Surprisingly, parity and mean-payoff MDPs share two non-trivial properties: they both have pure stationary optimal strategies [4, 15] and they both are approximable by discounted MDPs with multiple discount factors (multidiscounted MDPs) [5, 15] . In this paper we unify and generalize these results. We introduce a new class of payoff functions called the priority weighted payoff functions, which are generalization of both parity and mean-payoff functions. We prove that priority weighted MDPs admit optimal strategies that are pure and stationary, and that the priority weighted value of an MDP is the limit of the multi-discounted value when discount factors tend to 0 simultaneously at various speeds.
doi:10.1109/lics.2007.28 dblp:conf/lics/GimbertZ07 fatcat:7i3aoegw6bd4hkz7rg4n3b2yya