论文信息 - Unconstrained Influence Diagrams

Unconstrained Influence Diagrams

We extend the language of influence diagrams to cope with decision scenarios where the order of decisions and observations is not determined. As the ordering of decisions is dependent on the evidence, a step-strategy of such a scenario is a sequence of dependent choices of the next action. A strategy is a step-strategy together with selection functions for decision actions. The structure of a step-strategy can be represented as a DAG with nodes labeled with action variables. We introduce the concept of GS-DAG: a DAG incurporating an optimal step-strategy for any instantiation. We give a method for constructing GS-DAGs, and we show how to use a GS-DAG for determining an optimal strategy. Finally we discuss how analysis of relevant past can be used to reduce the size of the GS-DAG.

Finn Verner Jensen | Marta Vomlelová | F. V. Jensen | M. Vomlelová

[1] Anders L. Madsen,et al. Lazy Evaluation of Symmetric Bayesian Decision Problems , 1999, UAI.

[2] Prakash P. Shenoy,et al. Valuation network representation and solution of asymmetric decision problems , 2000, Eur. J. Oper. Res..

[3] Thomas D. Nielsen,et al. Representing and Solving Asymmetric Bayesian Decision Problems , 2000, UAI.

[4] Ross D. Shachter. Evaluating Influence Diagrams , 1986, Oper. Res..

[5] Steffen L. Lauritzen,et al. Representing and Solving Decision Problems with Limited Information , 2001, Manag. Sci..

[6] Thomas D. Nielsen,et al. Welldefined Decision Scenarios , 1999, UAI.

[7] Frank Jensen,et al. From Influence Diagrams to junction Trees , 1994, UAI.

[8] Prakash P. Shenoy,et al. Valuation-Based Systems for Bayesian Decision Analysis , 1992, Oper. Res..

[9] R. M. Oliver,et al. Representation and solution of decision problems using sequential decision diagrams , 1995 .

[10] Ross D. Shachter. Bayes-Ball: The Rational Pastime (for Determining Irrelevance and Requisite Information in Belief Networks and Influence Diagrams) , 1998, UAI.