Solving Influence Diagrams Using Gibbs Sampling
We describe a Monte Carlo method for solving influence diagrams. This method is a combination of stochastic dynamic programming and Gibbs sampling, an iterative Markov chain Monte Carlo method. Our method is especially useful when exact methods for solving influence diagrams fail.
Unable to display preview. Download preview PDF.
- Clemen, R.T. (1991) Making Hard Decisions, Duxbury Press, Belmont, California.Google Scholar
- Ndilikilikesha, P. (1992) Potential influence diagrams, Working Paper No. 235, School of Business, University of Kansas, Lawrence, Kansas.Google Scholar
- Olmsted, S. M. (1983) On representing and solving decision problems, Ph.D. thesis, Department of Engineering-Economic Systems, Stanford University, Stanford, CA.Google Scholar
- Shenoy, P. P. (1993) A new method for representing and solving Bayesian decision problems, in D. J. Hand (ed.), Artificial Intelligence Frontiers in Statistics: Al and Statistics III, 119–138, Chapman & Hall, London.Google Scholar