Strongly Polynomial Algorithms for Transient and Average-Cost MDPs

Open Webpage

Eugene A. Feinberg, Jefferson Huang

Published: 2017, Last Modified: 01 Oct 2024SIGMETRICS Perform. Evaluation Rev. 2017EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: This paper considers transient total-cost MDPs with transition rates whose values may be greater than one, and average-cost MDPs satisfying the condition that the expected time to hit a certain state from any initial state and under any stationary policy is bounded above by a constant. Linear programming formulations for such MDPs are provided that are solvable in strongly polynomial time.