Causal Graph Learning via Distributional Invariance of Cause-Effect Relationship

Nang Hung Nguyen, Trong Nghia Hoang, Thao Nguyen Truong, Phi Le Nguyen, Masashi Sugiyama

Keywords: Causal Graph Learning, Invariance

Abstract: This paper introduces a new framework for recovering causal graphs from observational data, leveraging the fact that the distribution of an effect, conditioned on its causes, remains invariant to changes in the prior distribution of those causes. This insight enables a direct test for potential causal relationships by checking the variance of their corresponding effect-cause conditional distributions across multiple downsampled subsets of the data. These subsets are selected to reflect different prior cause distributions, while preserving the effect-cause conditional relationships. Using this invariance test and exploiting an (empirical) sparsity of most causal graphs, we develop an algorithm that efficiently uncovers causal relationships with quadratic complexity in the number of observational features/variables, reducing the processing time by up to 25x compared to state-of-the-art methods. Our empirical studies on a diverse benchmark of large-scale datasets demonstrate that the developed algorithm consistently performs better or comparable to existing works while generally achieving better scalability.

Primary Area: causal reasoning

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 9437