Dynamic Programming in in Uence Diagrams with Decision Circuits
Abstract
Decision circuits perform efficient evaluation of influence diagrams, building on the advances in arithmetic circuits for belief network inference [Darwiche, 2003; Bhattacharjya and Shachter, 2007]. We show how even more compact decision circuits can be constructed for dynamic programming in influence diagrams with separable value functions and conditionally independent subproblems. Once a decision circuit has been constructed based on the diagram's "global" graphical structure, it can be compiled to exploit "local" structure for efficient evaluation and sensitivity analysis.
Cite
Text
Shachter and Bhattacharjya. "Dynamic Programming in in Uence Diagrams with Decision Circuits." Conference on Uncertainty in Artificial Intelligence, 2010.Markdown
[Shachter and Bhattacharjya. "Dynamic Programming in in Uence Diagrams with Decision Circuits." Conference on Uncertainty in Artificial Intelligence, 2010.](https://mlanthology.org/uai/2010/shachter2010uai-dynamic/)BibTeX
@inproceedings{shachter2010uai-dynamic,
title = {{Dynamic Programming in in Uence Diagrams with Decision Circuits}},
author = {Shachter, Ross D. and Bhattacharjya, Debarun},
booktitle = {Conference on Uncertainty in Artificial Intelligence},
year = {2010},
pages = {509-516},
url = {https://mlanthology.org/uai/2010/shachter2010uai-dynamic/}
}