Synonyms Relational Dynamic Programming , Dynamic Programming for Relational Domains , Relational Value Iteration Definition

Decision-theoretic planning aims at constructing a policy for acting in an uncertain environment that maximizes an agent’s expected utility along a sequence of steps. For this task, Markov decision processes (MDPs) have become the standard model. However, classical dynamic programming algorithms for solving MDPs require explicit state and action enumeration, which is often impractical: the number of states and actions grows very quickly with the number of domain objects and relations. In contrast, symbolic dynamic programming (SDP) algorithms seek to avoid explicit state and action enumeration through the symbolic representation of an MDP and a corresponding symbolic derivation of its solution, such as a value function. In essence, SDP algorithms exploit the symbolic structure of the MDP representation to construct a minimal logical partition of the state space required to make all necessary value distinctions.