Causality with Gates

John Winn
Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics, PMLR 22:1314-1322, 2012.

Abstract

An intervention on a variable removes the influences that usually have a causal effect on that variable. Gates are a general-purpose graphical modelling notation for representing such context-specific independencies in the structure of a graphical model. We extend d-separation to cover gated graphical models and show that it subsumes do calculus when gates are used to represent interventions. We also show how standard message passing inference algorithms, such as belief propagation, can be applied to the gated graph. This demonstrates that causal reasoning can be performed by probabilistic inference alone.

Cite this Paper

BibTeX
@InProceedings{pmlr-v22-winn12, title = {Causality with Gates}, author = {Winn, John}, booktitle = {Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics}, pages = {1314--1322}, year = {2012}, editor = {Lawrence, Neil D. and Girolami, Mark}, volume = {22}, series = {Proceedings of Machine Learning Research}, address = {La Palma, Canary Islands}, month = {21--23 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v22/winn12/winn12.pdf}, url = {https://proceedings.mlr.press/v22/winn12.html}, abstract = {An intervention on a variable removes the influences that usually have a causal effect on that variable. Gates are a general-purpose graphical modelling notation for representing such context-specific independencies in the structure of a graphical model. We extend d-separation to cover gated graphical models and show that it subsumes do calculus when gates are used to represent interventions. We also show how standard message passing inference algorithms, such as belief propagation, can be applied to the gated graph. This demonstrates that causal reasoning can be performed by probabilistic inference alone.} }
Endnote
%0 Conference Paper %T Causality with Gates %A John Winn %B Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2012 %E Neil D. Lawrence %E Mark Girolami %F pmlr-v22-winn12 %I PMLR %P 1314--1322 %U https://proceedings.mlr.press/v22/winn12.html %V 22 %X An intervention on a variable removes the influences that usually have a causal effect on that variable. Gates are a general-purpose graphical modelling notation for representing such context-specific independencies in the structure of a graphical model. We extend d-separation to cover gated graphical models and show that it subsumes do calculus when gates are used to represent interventions. We also show how standard message passing inference algorithms, such as belief propagation, can be applied to the gated graph. This demonstrates that causal reasoning can be performed by probabilistic inference alone.
RIS
TY - CPAPER TI - Causality with Gates AU - John Winn BT - Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics DA - 2012/03/21 ED - Neil D. Lawrence ED - Mark Girolami ID - pmlr-v22-winn12 PB - PMLR DP - Proceedings of Machine Learning Research VL - 22 SP - 1314 EP - 1322 L1 - http://proceedings.mlr.press/v22/winn12/winn12.pdf UR - https://proceedings.mlr.press/v22/winn12.html AB - An intervention on a variable removes the influences that usually have a causal effect on that variable. Gates are a general-purpose graphical modelling notation for representing such context-specific independencies in the structure of a graphical model. We extend d-separation to cover gated graphical models and show that it subsumes do calculus when gates are used to represent interventions. We also show how standard message passing inference algorithms, such as belief propagation, can be applied to the gated graph. This demonstrates that causal reasoning can be performed by probabilistic inference alone. ER -
APA
Winn, J.. (2012). Causality with Gates. Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 22:1314-1322 Available from https://proceedings.mlr.press/v22/winn12.html.

Related Material