Beyond Structural Causal Models: Causal Constraints Models

Tineke Blom, Stephan Bongers, Joris M. Mooij
Proceedings of The 35th Uncertainty in Artificial Intelligence Conference, PMLR 115:585-594, 2020.

Abstract

Structural Causal Models (SCMs) provide a popular causal modeling framework. In this work, we show that SCMs are not flexible enough to give a complete causal representation of dynamical systems at equilibrium. Instead, we propose a generalization of the notion of an SCM, that we call Causal Constraints Model (CCM), and prove that CCMs do capture the causal semantics of such systems. We show how CCMs can be constructed from differential equations and initial conditions and we illustrate our ideas further on a simple but ubiquitous (bio)chemical reaction. Our framework also allows to model functional laws, such as the ideal gas law, in a sensible and intuitive way.

Cite this Paper

BibTeX
@InProceedings{pmlr-v115-blom20a, title = {Beyond Structural Causal Models: Causal Constraints Models}, author = {Blom, Tineke and Bongers, Stephan and Mooij, Joris M.}, booktitle = {Proceedings of The 35th Uncertainty in Artificial Intelligence Conference}, pages = {585--594}, year = {2020}, editor = {Adams, Ryan P. and Gogate, Vibhav}, volume = {115}, series = {Proceedings of Machine Learning Research}, month = {22--25 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v115/blom20a/blom20a.pdf}, url = {https://proceedings.mlr.press/v115/blom20a.html}, abstract = {Structural Causal Models (SCMs) provide a popular causal modeling framework. In this work, we show that SCMs are not flexible enough to give a complete causal representation of dynamical systems at equilibrium. Instead, we propose a generalization of the notion of an SCM, that we call Causal Constraints Model (CCM), and prove that CCMs do capture the causal semantics of such systems. We show how CCMs can be constructed from differential equations and initial conditions and we illustrate our ideas further on a simple but ubiquitous (bio)chemical reaction. Our framework also allows to model functional laws, such as the ideal gas law, in a sensible and intuitive way.} }
Endnote
%0 Conference Paper %T Beyond Structural Causal Models: Causal Constraints Models %A Tineke Blom %A Stephan Bongers %A Joris M. Mooij %B Proceedings of The 35th Uncertainty in Artificial Intelligence Conference %C Proceedings of Machine Learning Research %D 2020 %E Ryan P. Adams %E Vibhav Gogate %F pmlr-v115-blom20a %I PMLR %P 585--594 %U https://proceedings.mlr.press/v115/blom20a.html %V 115 %X Structural Causal Models (SCMs) provide a popular causal modeling framework. In this work, we show that SCMs are not flexible enough to give a complete causal representation of dynamical systems at equilibrium. Instead, we propose a generalization of the notion of an SCM, that we call Causal Constraints Model (CCM), and prove that CCMs do capture the causal semantics of such systems. We show how CCMs can be constructed from differential equations and initial conditions and we illustrate our ideas further on a simple but ubiquitous (bio)chemical reaction. Our framework also allows to model functional laws, such as the ideal gas law, in a sensible and intuitive way.
APA
Blom, T., Bongers, S. & Mooij, J.M.. (2020). Beyond Structural Causal Models: Causal Constraints Models. Proceedings of The 35th Uncertainty in Artificial Intelligence Conference, in Proceedings of Machine Learning Research 115:585-594 Available from https://proceedings.mlr.press/v115/blom20a.html.

Related Material