Automated Identification of Causal Moderators in Time-Series Data

Min Zheng, Jan Claassen, Samantha Kleinberg
Proceedings of 2018 ACM SIGKDD Workshop on Causal Disocvery, PMLR 92:4-22, 2018.

Abstract

Causal inference is often taken to mean finding links between individual variables. However in many real-world cases, such as in biological systems, relationships are more complex, with groups of factors needed to produce an effect, or some factors only modifying other relationships rather than producing outcomes alone. For instance, weight may alter the efficacy of a drug without causing side effects itself. Such moderating factors may change the timing, intensity, or probability of a causal relationship. Distinguishing moderators from genuine causes can lead to more effective medical interventions, and better strategies for bringing about a desired effect, since a moderator alone is ineffective. However, there have not yet been algorithms to automatically infer moderators in a large-scale automated way, and they cannot be easily read off from causal graphs. We introduce a set of temporal logic rules to automatically identify the asymmetric roles of causes and moderators in a computationally efficient manner. Experiments on simulated data demonstrate that even in challenging cases we can find moderators and avoid confounding, and on real neurological ICU data we show how the approach can find more descriptive and meaningful relationships than the state of the art.

Cite this Paper

BibTeX
@InProceedings{pmlr-v92-zheng18a, title = {Automated Identification of Causal Moderators in Time-Series Data}, author = {Zheng, Min and Claassen, Jan and Kleinberg, Samantha}, booktitle = {Proceedings of 2018 ACM SIGKDD Workshop on Causal Disocvery}, pages = {4--22}, year = {2018}, editor = {Le, Thuc Duy and Zhang, Kun and Kıcıman, Emre and Hyvärinen, Aapo and Liu, Lin}, volume = {92}, series = {Proceedings of Machine Learning Research}, month = {20 Aug}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v92/zheng18a/zheng18a.pdf}, url = {https://proceedings.mlr.press/v92/zheng18a.html}, abstract = {Causal inference is often taken to mean finding links between individual variables. However in many real-world cases, such as in biological systems, relationships are more complex, with groups of factors needed to produce an effect, or some factors only modifying other relationships rather than producing outcomes alone. For instance, weight may alter the efficacy of a drug without causing side effects itself. Such moderating factors may change the timing, intensity, or probability of a causal relationship. Distinguishing moderators from genuine causes can lead to more effective medical interventions, and better strategies for bringing about a desired effect, since a moderator alone is ineffective. However, there have not yet been algorithms to automatically infer moderators in a large-scale automated way, and they cannot be easily read off from causal graphs. We introduce a set of temporal logic rules to automatically identify the asymmetric roles of causes and moderators in a computationally efficient manner. Experiments on simulated data demonstrate that even in challenging cases we can find moderators and avoid confounding, and on real neurological ICU data we show how the approach can find more descriptive and meaningful relationships than the state of the art.} }
Endnote
%0 Conference Paper %T Automated Identification of Causal Moderators in Time-Series Data %A Min Zheng %A Jan Claassen %A Samantha Kleinberg %B Proceedings of 2018 ACM SIGKDD Workshop on Causal Disocvery %C Proceedings of Machine Learning Research %D 2018 %E Thuc Duy Le %E Kun Zhang %E Emre Kıcıman %E Aapo Hyvärinen %E Lin Liu %F pmlr-v92-zheng18a %I PMLR %P 4--22 %U https://proceedings.mlr.press/v92/zheng18a.html %V 92 %X Causal inference is often taken to mean finding links between individual variables. However in many real-world cases, such as in biological systems, relationships are more complex, with groups of factors needed to produce an effect, or some factors only modifying other relationships rather than producing outcomes alone. For instance, weight may alter the efficacy of a drug without causing side effects itself. Such moderating factors may change the timing, intensity, or probability of a causal relationship. Distinguishing moderators from genuine causes can lead to more effective medical interventions, and better strategies for bringing about a desired effect, since a moderator alone is ineffective. However, there have not yet been algorithms to automatically infer moderators in a large-scale automated way, and they cannot be easily read off from causal graphs. We introduce a set of temporal logic rules to automatically identify the asymmetric roles of causes and moderators in a computationally efficient manner. Experiments on simulated data demonstrate that even in challenging cases we can find moderators and avoid confounding, and on real neurological ICU data we show how the approach can find more descriptive and meaningful relationships than the state of the art.
APA
Zheng, M., Claassen, J. & Kleinberg, S.. (2018). Automated Identification of Causal Moderators in Time-Series Data. Proceedings of 2018 ACM SIGKDD Workshop on Causal Disocvery, in Proceedings of Machine Learning Research 92:4-22 Available from https://proceedings.mlr.press/v92/zheng18a.html.

Related Material