STRUCTURING CONDITIONAL RELATIONSHIPS IN INFLUENCE DIAGRAMS

An influence diagram is a graphical representation of a decision probl em that is at once a formal description of a decision problem that can be treated by computers and a representation that is easily understoo d by decision makers who may be unskilled in the art of complex probab ilistic modeling. The power of an influence diagram, both as an analys is tool and a communication tool, lies in its ability to concisely sum marize the structure of a decision problem. However, when confronted w ith highly asymmetric problems in which particular acts or events lead to very different possibilities, many analysts prefer decision trees to influence diagrams. In this paper, we extend the definition of an i nfluence diagram by introducing a new representation for its condition al probability distributions. This extended influence diagram represen tation, combining elements of the decision tree and influence diagram representations, allows one to clearly and efficiently represent asymm etric decision problems and provides an attractive alternative to both the decision tree and conventional influence diagram representations.