To Predict or to Reject: Causal Effect Estimation with Uncertainty on Networked Data

TL;DR

This paper introduces GraphDKL, a novel framework that models uncertainty in causal effect estimation on networked data, addressing positivity violations and improving reliability of individual treatment effect predictions.

Contribution

The paper presents the first framework to incorporate uncertainty modeling and positivity violation handling in causal effect estimation on graph-structured data.

Findings

GraphDKL effectively identifies unreliable estimations.

The method outperforms existing approaches in accuracy.

Uncertainty modeling improves trustworthiness of causal predictions.

Abstract

Due to the imbalanced nature of networked observational data, the causal effect predictions for some individuals can severely violate the positivity/overlap assumption, rendering unreliable estimations. Nevertheless, this potential risk of individual-level treatment effect estimation on networked data has been largely under-explored. To create a more trustworthy causal effect estimator, we propose the uncertainty-aware graph deep kernel learning (GraphDKL) framework with Lipschitz constraint to model the prediction uncertainty with Gaussian process and identify unreliable estimations. To the best of our knowledge, GraphDKL is the first framework to tackle the violation of positivity assumption when performing causal effect estimation with graphs. With extensive experiments, we demonstrate the superiority of our proposed method in uncertainty-aware causal effect estimation on networked data.