A POMDP Approach to Influence Diagram Evaluation
Abstract
We propose a node-removal/arc-reversal algorithm for influence diagram evaluation that includes reductions that allow an influence diagram to be solved by a generalization of the dynamic programming approach to solving partially observable Markov decision processes (POMDPs). Among its potential advantages, the algorithm allows a more flexible ordering of node removals, and a POMDP-inspired approach to optimizing over hidden state variables, which can improve the scalability of influence diagram evaluation in solving complex, multi-stage problems. It also finds a more compact representation of an optimal strategy. PDF
Cite
Text
Hansen et al. "A POMDP Approach to Influence Diagram Evaluation." International Joint Conference on Artificial Intelligence, 2016.Markdown
[Hansen et al. "A POMDP Approach to Influence Diagram Evaluation." International Joint Conference on Artificial Intelligence, 2016.](https://mlanthology.org/ijcai/2016/hansen2016ijcai-pomdp/)BibTeX
@inproceedings{hansen2016ijcai-pomdp,
title = {{A POMDP Approach to Influence Diagram Evaluation}},
author = {Hansen, Eric A. and Shi, Jinchuan and Khaled, Arindam},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {2016},
pages = {3124-3132},
url = {https://mlanthology.org/ijcai/2016/hansen2016ijcai-pomdp/}
}