Solving Limited Memory Influence Diagrams

Fast local search methods for solving limited memory influence diagrams

Solving influence diagrams using heuristic search

Existing methods for solving influence diagrams are mostly based on the bottom-up dynamic programming technique. These methods may waste computation in solving decision scenarios that have zero probabilities or are unreachable from any initial state by following an optimal decision policy. Heuristic search was applied in (Qi & Poole 1995) to address these limitations, but their algorithm uses a...

On the complexity of solving polytree-shaped limited memory influence diagrams with binary variables

Influence diagrams are intuitive and concise representations of structured decision problems. When the problem is non-Markovian, an optimal strategy can be exponentially large in the size of the diagram. We can avoid the inherent intractability by constraining the size of admissible strategies, giving rise to limited memory influence diagrams. A valuable question is then how small do strategies...

Solving Limited Memory Influence Diagrams Using Branch-and-Bound Search

A limited-memory influence diagram (LIMID) is a generalization of a traditional influence diagram in which the assumptions of regularity and no-forgetting are relaxed. Lauritzen and Nilsson (2001) introduced this model and proposed an iterative algorithm, called single policy updating (SPU), that converges to a locally optimal solution. In this paper, we describe a branch-andbound algorithm tha...

Markov Limid processes for representing and solving renewal problems

In this paper a new tool for simultaneous optimization of decisions on multiple time scales is discussed. It combines the dynamic properties of Markov decision processes with the flexible and compact state space representation of Limited Memory Influence Diagrams (LIMIDs).