Decisions and Dependence in Influence Diagrams

Ross D. Shachter
Proceedings of the Eighth International Conference on Probabilistic Graphical Models, PMLR 52:462-473, 2016.

Abstract

The concept of dependence among variables in a Bayesian belief network is well understood, but what does it mean in an influence diagram where some of those variables are decisions? There are three quite different answers to this question that take the familiar concepts for uncertain variables and extend them to decisions. First is responsiveness, whether the choice for a decision affects another variable. Second is materiality, whether observing other variables before making a decision affects the choice for the decision and thus improves its quality. Third is the usual notion of dependence, assuming that all of the decisions are made optimally given the information available at the time of the decisions. There are some subtleties involved, but all three types of decision dependence can be quite useful for understanding a decision model.

Cite this Paper

BibTeX
@InProceedings{pmlr-v52-shachter16, title = {Decisions and Dependence in Influence Diagrams}, author = {Shachter, Ross D.}, booktitle = {Proceedings of the Eighth International Conference on Probabilistic Graphical Models}, pages = {462--473}, year = {2016}, editor = {Antonucci, Alessandro and Corani, Giorgio and Campos}, Cassio Polpo}, volume = {52}, series = {Proceedings of Machine Learning Research}, address = {Lugano, Switzerland}, month = {06--09 Sep}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v52/shachter16.pdf}, url = {https://proceedings.mlr.press/v52/shachter16.html}, abstract = {The concept of dependence among variables in a Bayesian belief network is well understood, but what does it mean in an influence diagram where some of those variables are decisions? There are three quite different answers to this question that take the familiar concepts for uncertain variables and extend them to decisions. First is responsiveness, whether the choice for a decision affects another variable. Second is materiality, whether observing other variables before making a decision affects the choice for the decision and thus improves its quality. Third is the usual notion of dependence, assuming that all of the decisions are made optimally given the information available at the time of the decisions. There are some subtleties involved, but all three types of decision dependence can be quite useful for understanding a decision model.} }
Endnote
%0 Conference Paper %T Decisions and Dependence in Influence Diagrams %A Ross D. Shachter %B Proceedings of the Eighth International Conference on Probabilistic Graphical Models %C Proceedings of Machine Learning Research %D 2016 %E Alessandro Antonucci %E Giorgio Corani %E Cassio Polpo Campos} %F pmlr-v52-shachter16 %I PMLR %P 462--473 %U https://proceedings.mlr.press/v52/shachter16.html %V 52 %X The concept of dependence among variables in a Bayesian belief network is well understood, but what does it mean in an influence diagram where some of those variables are decisions? There are three quite different answers to this question that take the familiar concepts for uncertain variables and extend them to decisions. First is responsiveness, whether the choice for a decision affects another variable. Second is materiality, whether observing other variables before making a decision affects the choice for the decision and thus improves its quality. Third is the usual notion of dependence, assuming that all of the decisions are made optimally given the information available at the time of the decisions. There are some subtleties involved, but all three types of decision dependence can be quite useful for understanding a decision model.
RIS
TY - CPAPER TI - Decisions and Dependence in Influence Diagrams AU - Ross D. Shachter BT - Proceedings of the Eighth International Conference on Probabilistic Graphical Models DA - 2016/08/15 ED - Alessandro Antonucci ED - Giorgio Corani ED - Cassio Polpo Campos} ID - pmlr-v52-shachter16 PB - PMLR DP - Proceedings of Machine Learning Research VL - 52 SP - 462 EP - 473 L1 - http://proceedings.mlr.press/v52/shachter16.pdf UR - https://proceedings.mlr.press/v52/shachter16.html AB - The concept of dependence among variables in a Bayesian belief network is well understood, but what does it mean in an influence diagram where some of those variables are decisions? There are three quite different answers to this question that take the familiar concepts for uncertain variables and extend them to decisions. First is responsiveness, whether the choice for a decision affects another variable. Second is materiality, whether observing other variables before making a decision affects the choice for the decision and thus improves its quality. Third is the usual notion of dependence, assuming that all of the decisions are made optimally given the information available at the time of the decisions. There are some subtleties involved, but all three types of decision dependence can be quite useful for understanding a decision model. ER -
APA
Shachter, R.D.. (2016). Decisions and Dependence in Influence Diagrams. Proceedings of the Eighth International Conference on Probabilistic Graphical Models, in Proceedings of Machine Learning Research 52:462-473 Available from https://proceedings.mlr.press/v52/shachter16.html.

Related Material