Causal Discovery with Multi-Domain LiNGAM for Latent Factors

Yan Zeng, Shohei Shimizu, Ruichu Cai, Feng Xie, Michio Yamamoto, Zhifeng Hao
Proceedings of The 2021 Causal Analysis Workshop Series, PMLR 160:1-4, 2021.

Abstract

Discovering causal structures among latent factors from observed data is particularly significant yet challenging problem. Despite some efforts for this problem, existing methods focus on the single-domain data only. In this paper, we propose Multi-Domain Linear Non-Gaussian Acyclic Models for LAtent Factors (MD-LiNA), where the causal structures from different domains may be different, but they have a shared causal structure among latent factors of interest. The model enriches the causal representation for multi-domain data. We propose an integrated two-phase algorithm to estimate the model. In particular, we first locate the latent factors and estimate the factor loading matrix. Then to uncover the shared causal structure among latent factors of interest, we derive a score function based on the characterization of independence relations between external influences and the dependence relations between multi-domain latent factors and latent factors of interest. Experimental results on synthetic data demonstrate the efficacy of our approach.

Cite this Paper


BibTeX
@InProceedings{pmlr-v160-zeng21a, title = {Causal Discovery with Multi-Domain LiNGAM for Latent Factors}, author = {Zeng, Yan and Shimizu, Shohei and Cai, Ruichu and Xie, Feng and Yamamoto, Michio and Hao, Zhifeng}, booktitle = {Proceedings of The 2021 Causal Analysis Workshop Series}, pages = {1--4}, year = {2021}, editor = {Ma, Sisi and Kummerfeld, Erich}, volume = {160}, series = {Proceedings of Machine Learning Research}, month = {16 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v160/zeng21a/zeng21a.pdf}, url = {https://proceedings.mlr.press/v160/zeng21a.html}, abstract = {Discovering causal structures among latent factors from observed data is particularly significant yet challenging problem. Despite some efforts for this problem, existing methods focus on the single-domain data only. In this paper, we propose Multi-Domain Linear Non-Gaussian Acyclic Models for LAtent Factors (MD-LiNA), where the causal structures from different domains may be different, but they have a shared causal structure among latent factors of interest. The model enriches the causal representation for multi-domain data. We propose an integrated two-phase algorithm to estimate the model. In particular, we first locate the latent factors and estimate the factor loading matrix. Then to uncover the shared causal structure among latent factors of interest, we derive a score function based on the characterization of independence relations between external influences and the dependence relations between multi-domain latent factors and latent factors of interest. Experimental results on synthetic data demonstrate the efficacy of our approach.} }
Endnote
%0 Conference Paper %T Causal Discovery with Multi-Domain LiNGAM for Latent Factors %A Yan Zeng %A Shohei Shimizu %A Ruichu Cai %A Feng Xie %A Michio Yamamoto %A Zhifeng Hao %B Proceedings of The 2021 Causal Analysis Workshop Series %C Proceedings of Machine Learning Research %D 2021 %E Sisi Ma %E Erich Kummerfeld %F pmlr-v160-zeng21a %I PMLR %P 1--4 %U https://proceedings.mlr.press/v160/zeng21a.html %V 160 %X Discovering causal structures among latent factors from observed data is particularly significant yet challenging problem. Despite some efforts for this problem, existing methods focus on the single-domain data only. In this paper, we propose Multi-Domain Linear Non-Gaussian Acyclic Models for LAtent Factors (MD-LiNA), where the causal structures from different domains may be different, but they have a shared causal structure among latent factors of interest. The model enriches the causal representation for multi-domain data. We propose an integrated two-phase algorithm to estimate the model. In particular, we first locate the latent factors and estimate the factor loading matrix. Then to uncover the shared causal structure among latent factors of interest, we derive a score function based on the characterization of independence relations between external influences and the dependence relations between multi-domain latent factors and latent factors of interest. Experimental results on synthetic data demonstrate the efficacy of our approach.
APA
Zeng, Y., Shimizu, S., Cai, R., Xie, F., Yamamoto, M. & Hao, Z.. (2021). Causal Discovery with Multi-Domain LiNGAM for Latent Factors. Proceedings of The 2021 Causal Analysis Workshop Series, in Proceedings of Machine Learning Research 160:1-4 Available from https://proceedings.mlr.press/v160/zeng21a.html.

Related Material