Sample-Specific Root Causal Inference with Latent Variables

TL;DR

This paper introduces EEL, a new method for sample-specific root causal inference that accounts for latent confounders, improving accuracy and robustness without needing to estimate the full causal graph.

Contribution

The paper proposes EEL, an algorithm that relaxes assumptions of no latent confounding in root causal analysis, enabling more accurate identification of initial causes.

Findings

EEL outperforms previous methods in accuracy.

EEL is robust to latent confounders.

EEL avoids full causal graph estimation.

Abstract

Root causal analysis seeks to identify the set of initial perturbations that induce an unwanted outcome. In prior work, we defined sample-specific root causes of disease using exogenous error terms that predict a diagnosis in a structural equation model. We rigorously quantified predictivity using Shapley values. However, the associated algorithms for inferring root causes assume no latent confounding. We relax this assumption by permitting confounding among the predictors. We then introduce a corresponding procedure called Extract Errors with Latents (EEL) for recovering the error terms up to contamination by vertices on certain paths under the linear non-Gaussian acyclic model. EEL also identifies the smallest sets of dependent errors for fast computation of the Shapley values. The algorithm bypasses the hard problem of estimating the underlying causal graph in both cases. Experiments highlight the superior accuracy and robustness of EEL relative to its predecessors.