Learning Causal Structures Using Regression Invariance
arXiv:1705.09644
Abstract
We study causal inference in a multi-environment setting, in which the functional relations for producing the variables from their direct causes remain the same across environments, while the distribution of exogenous noises may vary. We introduce the idea of using the invariance of the functional relations of the variables to their causes across a set of environments. We define a notion of completeness for a causal inference algorithm in this setting and prove the existence of such algorithm by proposing the baseline algorithm. Additionally, we present an alternate algorithm that has significantly improved computational and sample complexity compared to the baseline algorithm. The experiment results show that the proposed algorithm outperforms the other existing algorithms.
References in corpus (6)
- Identifiability of Gaussian structural equation models with equal error variances
- Two Optimal Strategies for Active Learning of Causal Models from Interventional Data
- Causal Discovery from Changes
- Learning Causal Graphs with Small Interventions
- Optimal Experiment Design for Causal Discovery from Fixed Number of Experiments
- Interaction Information for Causal Inference: The Case of Directed Triangle
Cited by in corpus (7)
- Direct Estimation of Differences in Causal Graphs
- Proactive Pseudo-Intervention: Causally Informed Contrastive Learning For Interpretable Vision Models
- Causality-aware counterfactual confounding adjustment for feature representations learned by deep models
- Environment Invariant Linear Least Squares
- Improving Model Robustness Using Causal Knowledge
- Causal Discovery with Multi-Domain LiNGAM for Latent Factors
- Orthogonal Structure Search for Efficient Causal Discovery from Observational Data