From Influence Diagrams to Junction Trees
This work addresses decision-making under uncertainty for researchers and practitioners in probabilistic graphical models, but it appears incremental as it builds on existing influence diagram and junction tree techniques.
The paper tackles the problem of solving decision problems formulated as influence diagrams by introducing an approach that involves triangulation, junction tree construction, and message passing, resulting in a method for computing expected utilities and optimal decision policies.
We present an approach to the solution of decision problems formulated as influence diagrams. This approach involves a special triangulation of the underlying graph, the construction of a junction tree with special properties, and a message passing algorithm operating on the junction tree for computation of expected utilities and optimal decision policies.