A Forward Monte Carlo Method for Solving Influence Diagrams using local Computation

John M. Charnes, Prakash P. Shenoy
Proceedings of the Sixth International Workshop on Artificial Intelligence and Statistics, PMLR R1:75-82, 1997.

Abstract

The main goal of this paper is to describe a Monte Carlo method for solving influence diagrams using local computation. The forward Monte Carlo sampling technique draws independent and identically distributed observations. Methods that have been proposed in this spirit sample from the entire distribution. However, when the number of variables is large, the state space of all variables is exponentially large, and the sample size required for good estimates is too large to be practical. In the forward Monte Carlo method we generate observations from a subset of chance variables for each decision node in the influence diagram. We use methods developed for exact solution of influence diagrams to limit the number of chance variables sampled at any time. Because influence diagrams model each chance variable with a conditional probability distribution, the forward Monte Carlo solution method lends itself very well to influence-diagram representations.

Cite this Paper


BibTeX
@InProceedings{pmlr-vR1-charnes97a, title = {A Forward {M}onte {C}arlo Method for Solving Influence Diagrams using local Computation}, author = {Charnes, John M. and Shenoy, Prakash P.}, booktitle = {Proceedings of the Sixth International Workshop on Artificial Intelligence and Statistics}, pages = {75--82}, year = {1997}, editor = {Madigan, David and Smyth, Padhraic}, volume = {R1}, series = {Proceedings of Machine Learning Research}, month = {04--07 Jan}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/r1/charnes97a/charnes97a.pdf}, url = {https://proceedings.mlr.press/r1/charnes97a.html}, abstract = {The main goal of this paper is to describe a Monte Carlo method for solving influence diagrams using local computation. The forward Monte Carlo sampling technique draws independent and identically distributed observations. Methods that have been proposed in this spirit sample from the entire distribution. However, when the number of variables is large, the state space of all variables is exponentially large, and the sample size required for good estimates is too large to be practical. In the forward Monte Carlo method we generate observations from a subset of chance variables for each decision node in the influence diagram. We use methods developed for exact solution of influence diagrams to limit the number of chance variables sampled at any time. Because influence diagrams model each chance variable with a conditional probability distribution, the forward Monte Carlo solution method lends itself very well to influence-diagram representations.}, note = {Reissued by PMLR on 30 March 2021.} }
Endnote
%0 Conference Paper %T A Forward Monte Carlo Method for Solving Influence Diagrams using local Computation %A John M. Charnes %A Prakash P. Shenoy %B Proceedings of the Sixth International Workshop on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 1997 %E David Madigan %E Padhraic Smyth %F pmlr-vR1-charnes97a %I PMLR %P 75--82 %U https://proceedings.mlr.press/r1/charnes97a.html %V R1 %X The main goal of this paper is to describe a Monte Carlo method for solving influence diagrams using local computation. The forward Monte Carlo sampling technique draws independent and identically distributed observations. Methods that have been proposed in this spirit sample from the entire distribution. However, when the number of variables is large, the state space of all variables is exponentially large, and the sample size required for good estimates is too large to be practical. In the forward Monte Carlo method we generate observations from a subset of chance variables for each decision node in the influence diagram. We use methods developed for exact solution of influence diagrams to limit the number of chance variables sampled at any time. Because influence diagrams model each chance variable with a conditional probability distribution, the forward Monte Carlo solution method lends itself very well to influence-diagram representations. %Z Reissued by PMLR on 30 March 2021.
APA
Charnes, J.M. & Shenoy, P.P.. (1997). A Forward Monte Carlo Method for Solving Influence Diagrams using local Computation. Proceedings of the Sixth International Workshop on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research R1:75-82 Available from https://proceedings.mlr.press/r1/charnes97a.html. Reissued by PMLR on 30 March 2021.

Related Material