Nonlinear Causal Discovery with Latent Confounders

David Kaltenpoth, Jilles Vreeken
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:15639-15654, 2023.

Abstract

Causal discovery, the task of discovering the causal graph over a set of observed variables $X_1,\ldots,X_m$, is a challenging problem. One of the cornerstone assumptions is that of causal sufficiency: that all common causes of all measured variables have been observed. When it does not hold, causal discovery algorithms making this assumption return networks with many spurious edges. In this paper, we propose a nonlinear causal model involving hidden confounders. We show that it is identifiable from only the observed data and propose an efficient method for recovering this causal model. At the heart of our approach is a variational autoencoder which parametrizes both the causal interactions between observed variables as well as the influence of the unobserved confounders. Empirically we show that it outperforms other state-of-the-art methods for causal discovery under latent confounding on synthetic and real-world data.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-kaltenpoth23a, title = {Nonlinear Causal Discovery with Latent Confounders}, author = {Kaltenpoth, David and Vreeken, Jilles}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {15639--15654}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/kaltenpoth23a/kaltenpoth23a.pdf}, url = {https://proceedings.mlr.press/v202/kaltenpoth23a.html}, abstract = {Causal discovery, the task of discovering the causal graph over a set of observed variables $X_1,\ldots,X_m$, is a challenging problem. One of the cornerstone assumptions is that of causal sufficiency: that all common causes of all measured variables have been observed. When it does not hold, causal discovery algorithms making this assumption return networks with many spurious edges. In this paper, we propose a nonlinear causal model involving hidden confounders. We show that it is identifiable from only the observed data and propose an efficient method for recovering this causal model. At the heart of our approach is a variational autoencoder which parametrizes both the causal interactions between observed variables as well as the influence of the unobserved confounders. Empirically we show that it outperforms other state-of-the-art methods for causal discovery under latent confounding on synthetic and real-world data.} }
Endnote
%0 Conference Paper %T Nonlinear Causal Discovery with Latent Confounders %A David Kaltenpoth %A Jilles Vreeken %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-kaltenpoth23a %I PMLR %P 15639--15654 %U https://proceedings.mlr.press/v202/kaltenpoth23a.html %V 202 %X Causal discovery, the task of discovering the causal graph over a set of observed variables $X_1,\ldots,X_m$, is a challenging problem. One of the cornerstone assumptions is that of causal sufficiency: that all common causes of all measured variables have been observed. When it does not hold, causal discovery algorithms making this assumption return networks with many spurious edges. In this paper, we propose a nonlinear causal model involving hidden confounders. We show that it is identifiable from only the observed data and propose an efficient method for recovering this causal model. At the heart of our approach is a variational autoencoder which parametrizes both the causal interactions between observed variables as well as the influence of the unobserved confounders. Empirically we show that it outperforms other state-of-the-art methods for causal discovery under latent confounding on synthetic and real-world data.
APA
Kaltenpoth, D. & Vreeken, J.. (2023). Nonlinear Causal Discovery with Latent Confounders. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:15639-15654 Available from https://proceedings.mlr.press/v202/kaltenpoth23a.html.

Related Material