A Logical Characterization of Constraint-Based Causal Discovery

Tom Claassen, Tom Heskes
Proceedings of the 27th Conference on Uncertainty in Artificial Intelligence, PMLR R9:163-172, 2011.

Abstract

We present a novel approach to constraint-based causal discovery, that takes the form of straightforward logical inference, applied to a list of simple, logical statements about causal relations that are derived directly from observed (in)dependencies. It is both sound and complete, in the sense that all invariant features of the corresponding partial ancestral graph (PAG) are identified, even in the presence of latent variables and selection bias. The approach shows that every identifiable causal relation corresponds to one of just two fundamental forms. More importantly, as the basic building blocks of the method do not rely on the detailed (graphical) structure of the corresponding PAG, it opens up a range of new opportunities, including more robust inference, detailed accountability, and application to large models.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR9-claassen11a, title = {A Logical Characterization of Constraint-Based Causal Discovery}, author = {Claassen, Tom and Heskes, Tom}, booktitle = {Proceedings of the 27th Conference on Uncertainty in Artificial Intelligence}, pages = {163--172}, year = {2011}, editor = {Cozman, Fabio and Pfeffer, Avi}, volume = {R9}, series = {Proceedings of Machine Learning Research}, month = {14--17 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r9/main/assets/claassen11a/claassen11a.pdf}, url = {https://proceedings.mlr.press/r9/claassen11a.html}, abstract = {We present a novel approach to constraint-based causal discovery, that takes the form of straightforward logical inference, applied to a list of simple, logical statements about causal relations that are derived directly from observed (in)dependencies. It is both sound and complete, in the sense that all invariant features of the corresponding partial ancestral graph (PAG) are identified, even in the presence of latent variables and selection bias. The approach shows that every identifiable causal relation corresponds to one of just two fundamental forms. More importantly, as the basic building blocks of the method do not rely on the detailed (graphical) structure of the corresponding PAG, it opens up a range of new opportunities, including more robust inference, detailed accountability, and application to large models.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T A Logical Characterization of Constraint-Based Causal Discovery %A Tom Claassen %A Tom Heskes %B Proceedings of the 27th Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2011 %E Fabio Cozman %E Avi Pfeffer %F pmlr-vR9-claassen11a %I PMLR %P 163--172 %U https://proceedings.mlr.press/r9/claassen11a.html %V R9 %X We present a novel approach to constraint-based causal discovery, that takes the form of straightforward logical inference, applied to a list of simple, logical statements about causal relations that are derived directly from observed (in)dependencies. It is both sound and complete, in the sense that all invariant features of the corresponding partial ancestral graph (PAG) are identified, even in the presence of latent variables and selection bias. The approach shows that every identifiable causal relation corresponds to one of just two fundamental forms. More importantly, as the basic building blocks of the method do not rely on the detailed (graphical) structure of the corresponding PAG, it opens up a range of new opportunities, including more robust inference, detailed accountability, and application to large models. %Z Reissued by PMLR on 04 October 2026.
APA
Claassen, T. & Heskes, T.. (2011). A Logical Characterization of Constraint-Based Causal Discovery. Proceedings of the 27th Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R9:163-172 Available from https://proceedings.mlr.press/r9/claassen11a.html. Reissued by PMLR on 04 October 2026.

Related Material