Cyclic Causal Discovery from Continuous Equilibrium Data

Joris Mooij, Tom Heskes
Proceedings of the 29th Conference on Uncertainty in Artificial Intelligence, PMLR R11:175-183, 2013.

Abstract

We propose a method for learning cyclic causal models from a combination of obser- vational and interventional equilibrium data. Novel aspects of the proposed method are its ability to work with continuous data (without assuming linearity) and to deal with feedback loops. Within the context of biochemical re- actions, we also propose a novel way of mod- eling interventions that modify the activity of compounds instead of their abundance. For computational reasons, we approximate the nonlinear causal mechanisms by (coupled) lo- cal linearizations, one for each experimental condition. We apply the method to recon- struct a cellular signaling network from the flow cytometry data measured by Sachs et al. (2005). We show that our method finds evi- dence in the data for feedback loops and that it gives a more accurate quantitative descrip- tion of the data at comparable model com- plexity.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR11-mooij13a, title = {Cyclic Causal Discovery from Continuous Equilibrium Data}, author = {Mooij, Joris and Heskes, Tom}, booktitle = {Proceedings of the 29th Conference on Uncertainty in Artificial Intelligence}, pages = {175--183}, year = {2013}, editor = {Nicholson, Ann and Smyth, Padhraic}, volume = {R11}, series = {Proceedings of Machine Learning Research}, month = {12--14 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r11/main/assets/mooij13a/mooij13a.pdf}, url = {https://proceedings.mlr.press/r11/mooij13a.html}, abstract = {We propose a method for learning cyclic causal models from a combination of obser- vational and interventional equilibrium data. Novel aspects of the proposed method are its ability to work with continuous data (without assuming linearity) and to deal with feedback loops. Within the context of biochemical re- actions, we also propose a novel way of mod- eling interventions that modify the activity of compounds instead of their abundance. For computational reasons, we approximate the nonlinear causal mechanisms by (coupled) lo- cal linearizations, one for each experimental condition. We apply the method to recon- struct a cellular signaling network from the flow cytometry data measured by Sachs et al. (2005). We show that our method finds evi- dence in the data for feedback loops and that it gives a more accurate quantitative descrip- tion of the data at comparable model com- plexity.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T Cyclic Causal Discovery from Continuous Equilibrium Data %A Joris Mooij %A Tom Heskes %B Proceedings of the 29th Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2013 %E Ann Nicholson %E Padhraic Smyth %F pmlr-vR11-mooij13a %I PMLR %P 175--183 %U https://proceedings.mlr.press/r11/mooij13a.html %V R11 %X We propose a method for learning cyclic causal models from a combination of obser- vational and interventional equilibrium data. Novel aspects of the proposed method are its ability to work with continuous data (without assuming linearity) and to deal with feedback loops. Within the context of biochemical re- actions, we also propose a novel way of mod- eling interventions that modify the activity of compounds instead of their abundance. For computational reasons, we approximate the nonlinear causal mechanisms by (coupled) lo- cal linearizations, one for each experimental condition. We apply the method to recon- struct a cellular signaling network from the flow cytometry data measured by Sachs et al. (2005). We show that our method finds evi- dence in the data for feedback loops and that it gives a more accurate quantitative descrip- tion of the data at comparable model com- plexity. %Z Reissued by PMLR on 04 October 2026.
APA
Mooij, J. & Heskes, T.. (2013). Cyclic Causal Discovery from Continuous Equilibrium Data. Proceedings of the 29th Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R11:175-183 Available from https://proceedings.mlr.press/r11/mooij13a.html. Reissued by PMLR on 04 October 2026.

Related Material