| http://www.w3.org/ns/prov#value | - The main result is an algorithm which, given a decision process with non-Markovian rewards expressed in this manner, automatically constructs an equivalent MDP (with Markovian reward structure), allowing optimal policy construction using standard techniques.
|
