A Note on Dynamic Programming with Unbounded Rewards
In a recent paper, Lippman presents sufficient conditions for Denardo's N-stage contraction in discounted semi-Markov decision processes with unbounded rewards. In this note it is demonstrated that Lippman's conditions may be replaced by weaker conditions which even imply l-stage contraction. The verification of the conditions of this note is somewhat easier.
|Keywords||Markov processing, contraction operations, decision making, decision theory, dynamic programming, management science, mathematical anlysis, mathematical models|
van Nunen, J.A.E.E., & Wessels, J.. (1978). A Note on Dynamic Programming with Unbounded Rewards. Management Science, 24(5), 576–580. Retrieved from http://hdl.handle.net/1765/19890
|Permalink Final Version|