Cross-validating causal discovery via Leave-One-Variable-Out
Not on the map. Why?
Abstract (arXiv)
We propose a new approach to falsify causal discovery algorithms without ground truth, which is based on testing the causal model on a variable pair excluded during learning the causal model. Specifically, given data on X, Y, boldsymbolZ=X, Y, Z1,dots,Zk, we apply the causal discovery algorithm separately to the 'leave-one-out' data sets X, boldsymbolZ and Y, boldsymbolZ. We demonstrate that the two resulting causal models, in the form DAGs, ADMGs, CPDAGs or PAGs, often entail conclusions on the dependencies between X and Y and allow to estimate mathbbE(Ymid X=x) without any joint observations of X and Y, given only the leave-one-out datasets. This estimation is called "Leave-One-Variable-Out (LOVO)" prediction. Its error can be estimated since the joint distribution P(X, Y) is available, and X and Y have only been omitted for the purpose of falsification. We present two variants of LOVO prediction: One graphical method, which is applicable to general causal discovery algorithms, and one version tailored towards algorithms relying on specific a priori assumptions, such as linear additive noise models. Simulations indicate that the LOVO prediction error is indeed correlated with the accuracy of the causal outputs, affirming the method's effectiveness.
Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 19 of its 50 references that are in the map. First posted on arXiv 8 Nov 2024.
Builds on
19 of 50 references.
References · 19
- 2015Causal inference using invariant prediction: identification and confidence intervalscited by 536
- 2019Gradient-Based Neural DAG Learningcited by 227
- 2021Beware of the Simulated DAG! Causal Discovery Benchmarks May Be Easy To Gamecited by 161
- 2018Anchor regression: heterogeneous data meets causalitycited by 118
- 2022Score matching enables causal discovery of nonlinear additive noise modelscited by 111
- 2023Identifiability Guarantees for Causal Disentanglement from Soft Interventionscited by 84
- 2018Graphical Models for Processing Missing Datacited by 58
- 2013Causal Discovery from Changescited by 52
- 2023Scalable Causal Discovery with Score Matchingcited by 45
- 2019Learning Linear Non-Gaussian Causal Models in the Presence of Latent Variablescited by 39
- 2016Entropic Causal Inferencecited by 27
- 2022Causal Inference Through the Structural Causal Marginal Problemcited by 23
7 more
- 2023Self-Compatibility: Evaluating Causal Discovery without Ground Truthcited by 22
- 2019Causal discovery in heavy-tailed modelscited by 20
- 2022Deep Learning of Causal Structures in High Dimensionscited by 20
- 2019Integrating overlapping datasets using bivariate causal discoverycited by 12
- 2021Obtaining Causal Information by Merging Datasets with MAXENTcited by 9
- 2024Causal Discovery of Linear Non-Gaussian Causal Models with Unobserved Confoundingcited by 6
- 2023Out-of-Variable Generalization for Discriminative Modelscited by 4
Built on by
Citing papers per year: 2025: 1; 2026: 2.
3 papers.