arXiv Analytics

Sign in

arXiv:2102.13647 [stat.ML]AbstractReferencesReviewsResources

Beware of the Simulated DAG! Varsortability in Additive Noise Models

Alexander G. Reisach, Christof Seiler, Sebastian Weichwald

Published 2021-02-26Version 1

Additive noise models are a class of causal models in which each variable is defined as a function of its causes plus independent noise. In such models, the ordering of variables by marginal variances may be indicative of the causal order. We introduce varsortability as a measure of agreement between the ordering by marginal variance and the causal order. We show how varsortability dominates the performance of continuous structure learning algorithms on synthetic data. On real-world data, varsortability is an implausible and untestable assumption and we find no indication of high varsortability. We aim to raise awareness that varsortability easily occurs in simulated additive noise models. We provide a baseline method that explicitly exploits varsortability and advocate reporting varsortability in benchmarking data.

Related articles: Most relevant | Search more
arXiv:1205.2640 [stat.ML] (Published 2012-05-09)
Identifying confounders using additive noise models
arXiv:2108.11320 [stat.ML] (Published 2021-08-24)
The Effect of Noise Level on Causal Identification with Additive Noise Models
arXiv:2303.18211 [stat.ML] (Published 2023-03-31)
Simple Sorting Criteria Help Find the Causal Order in Additive Noise Models