Local Causal Discovery with Linear non-Gaussian Cyclic Models

Haoyue Dai, Ignavier Ng, Yujia Zheng, Zhengqing Gao, Kun Zhang
Proceedings of The 27th International Conference on Artificial Intelligence and Statistics, PMLR 238:154-162, 2024.

Abstract

Local causal discovery is of great practical significance, as there are often situations where the discovery of the global causal structure is unnecessary, and the interest lies solely on a single target variable. Most existing local methods utilize conditional independence relations, providing only a partially directed graph, and assume acyclicity for the ground-truth structure, even though real-world scenarios often involve cycles like feedback mechanisms. In this work, we present a general, unified local causal discovery method with linear non-Gaussian models, whether they are cyclic or acyclic. We extend the application of independent component analysis from the global context to independent subspace analysis, enabling the exact identification of the equivalent local directed structures and causal strengths from the Markov blanket of the target variable. We also propose an alternative regression-based method in the particular acyclic scenarios. Our identifiability results are empirically validated using both synthetic and real-world datasets.

Cite this Paper

BibTeX
@InProceedings{pmlr-v238-dai24a, title = {Local Causal Discovery with Linear non-{G}aussian Cyclic Models}, author = {Dai, Haoyue and Ng, Ignavier and Zheng, Yujia and Gao, Zhengqing and Zhang, Kun}, booktitle = {Proceedings of The 27th International Conference on Artificial Intelligence and Statistics}, pages = {154--162}, year = {2024}, editor = {Dasgupta, Sanjoy and Mandt, Stephan and Li, Yingzhen}, volume = {238}, series = {Proceedings of Machine Learning Research}, month = {02--04 May}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v238/dai24a/dai24a.pdf}, url = {https://proceedings.mlr.press/v238/dai24a.html}, abstract = {Local causal discovery is of great practical significance, as there are often situations where the discovery of the global causal structure is unnecessary, and the interest lies solely on a single target variable. Most existing local methods utilize conditional independence relations, providing only a partially directed graph, and assume acyclicity for the ground-truth structure, even though real-world scenarios often involve cycles like feedback mechanisms. In this work, we present a general, unified local causal discovery method with linear non-Gaussian models, whether they are cyclic or acyclic. We extend the application of independent component analysis from the global context to independent subspace analysis, enabling the exact identification of the equivalent local directed structures and causal strengths from the Markov blanket of the target variable. We also propose an alternative regression-based method in the particular acyclic scenarios. Our identifiability results are empirically validated using both synthetic and real-world datasets.} }
Endnote
%0 Conference Paper %T Local Causal Discovery with Linear non-Gaussian Cyclic Models %A Haoyue Dai %A Ignavier Ng %A Yujia Zheng %A Zhengqing Gao %A Kun Zhang %B Proceedings of The 27th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2024 %E Sanjoy Dasgupta %E Stephan Mandt %E Yingzhen Li %F pmlr-v238-dai24a %I PMLR %P 154--162 %U https://proceedings.mlr.press/v238/dai24a.html %V 238 %X Local causal discovery is of great practical significance, as there are often situations where the discovery of the global causal structure is unnecessary, and the interest lies solely on a single target variable. Most existing local methods utilize conditional independence relations, providing only a partially directed graph, and assume acyclicity for the ground-truth structure, even though real-world scenarios often involve cycles like feedback mechanisms. In this work, we present a general, unified local causal discovery method with linear non-Gaussian models, whether they are cyclic or acyclic. We extend the application of independent component analysis from the global context to independent subspace analysis, enabling the exact identification of the equivalent local directed structures and causal strengths from the Markov blanket of the target variable. We also propose an alternative regression-based method in the particular acyclic scenarios. Our identifiability results are empirically validated using both synthetic and real-world datasets.
APA
Dai, H., Ng, I., Zheng, Y., Gao, Z. & Zhang, K.. (2024). Local Causal Discovery with Linear non-Gaussian Cyclic Models. Proceedings of The 27th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 238:154-162 Available from https://proceedings.mlr.press/v238/dai24a.html.

Related Material