Learning the Structure of Causal Models with Relational and Temporal Dependence

Katerina Marazopoulou, Marc Maier, David Jensen
Proceedings of the 31st Conference on Uncertainty in Artificial Intelligence, PMLR R13:229-238, 2015.

Abstract

Many real-world domains are inherently relational and temporal—they consist of heterogeneous entities that interact with each over time. Effective reasoning about causality in such domains requires representations that explicitly model relational and temporal dependence. In this work, we provide a formalization of temporal relational models. We define temporal extensions to abstract ground graphs—a lifted representation that abstracts paths of dependence over all possible ground graphs. Temporal abstract ground graphs enable a sound and complete method for answering d-separation queries on temporal relational models. These methods provide the foundation for a constraint-based algorithm, TRCD, that learns causal models from temporal relational data. We provide experimental evidence that demonstrates the need to explicitly represent time when inferring causal dependence. We also demonstrate the expressive gain of TRCD compared to earlier algorithms that do not explicitly represent time.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR13-marazopoulou15a, title = {Learning the Structure of Causal Models with Relational and Temporal Dependence}, author = {Marazopoulou, Katerina and Maier, Marc and Jensen, David}, booktitle = {Proceedings of the 31st Conference on Uncertainty in Artificial Intelligence}, pages = {229--238}, year = {2015}, editor = {Meila, Marina and Heskes, Tom}, volume = {R13}, series = {Proceedings of Machine Learning Research}, month = {12--16 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r13/main/assets/marazopoulou15a/marazopoulou15a.pdf}, url = {https://proceedings.mlr.press/r13/marazopoulou15a.html}, abstract = {Many real-world domains are inherently relational and temporal—they consist of heterogeneous entities that interact with each over time. Effective reasoning about causality in such domains requires representations that explicitly model relational and temporal dependence. In this work, we provide a formalization of temporal relational models. We define temporal extensions to abstract ground graphs—a lifted representation that abstracts paths of dependence over all possible ground graphs. Temporal abstract ground graphs enable a sound and complete method for answering d-separation queries on temporal relational models. These methods provide the foundation for a constraint-based algorithm, TRCD, that learns causal models from temporal relational data. We provide experimental evidence that demonstrates the need to explicitly represent time when inferring causal dependence. We also demonstrate the expressive gain of TRCD compared to earlier algorithms that do not explicitly represent time.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T Learning the Structure of Causal Models with Relational and Temporal Dependence %A Katerina Marazopoulou %A Marc Maier %A David Jensen %B Proceedings of the 31st Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2015 %E Marina Meila %E Tom Heskes %F pmlr-vR13-marazopoulou15a %I PMLR %P 229--238 %U https://proceedings.mlr.press/r13/marazopoulou15a.html %V R13 %X Many real-world domains are inherently relational and temporal—they consist of heterogeneous entities that interact with each over time. Effective reasoning about causality in such domains requires representations that explicitly model relational and temporal dependence. In this work, we provide a formalization of temporal relational models. We define temporal extensions to abstract ground graphs—a lifted representation that abstracts paths of dependence over all possible ground graphs. Temporal abstract ground graphs enable a sound and complete method for answering d-separation queries on temporal relational models. These methods provide the foundation for a constraint-based algorithm, TRCD, that learns causal models from temporal relational data. We provide experimental evidence that demonstrates the need to explicitly represent time when inferring causal dependence. We also demonstrate the expressive gain of TRCD compared to earlier algorithms that do not explicitly represent time. %Z Reissued by PMLR on 04 October 2026.
APA
Marazopoulou, K., Maier, M. & Jensen, D.. (2015). Learning the Structure of Causal Models with Relational and Temporal Dependence. Proceedings of the 31st Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R13:229-238 Available from https://proceedings.mlr.press/r13/marazopoulou15a.html. Reissued by PMLR on 04 October 2026.

Related Material