In discrete-time Markov Decision Processes, decisions are made at discrete time epoch. However, for Continuous-time Markov Decision Process, decisions can be made at any time when decision maker wants. Unlike discrete-time Markov Decision Process, Continuous-time Markov Decision Process could better model the decision making process when the interested system has continuous dynamics, i.e., the system dynamics is defined by partial differential equations (PDEs).
Read more about this topic: Markov Decision Process
Famous quotes containing the words decision and/or process:
“A good decision is based on knowledge and not on numbers.”
—Plato (c. 427–347 B.C.)
“In contrast to revenge, which is the natural, automatic reaction to transgression and which, because of the irreversibility of the action process can be expected and even calculated, the act of forgiving can never be predicted; it is the only reaction that acts in an unexpected way and thus retains, though being a reaction, something of the original character of action.”
—Hannah Arendt (1906–1975)