VLUCI: Variational Learning of Unobserved Confounders for Counterfactual Inference

TL;DR

VLUCI introduces a variational learning framework to model unobserved confounders, enhancing the accuracy of counterfactual inference in observational studies by relaxing unconfoundedness assumptions.

Contribution

It presents a novel doubly variational inference model that explicitly captures unobserved confounders, improving counterfactual prediction accuracy over existing methods.

Findings

VLUCI outperforms existing models on synthetic datasets.

It provides reliable confidence intervals for counterfactual outcomes.

The model is compatible with state-of-the-art counterfactual inference techniques.

Abstract

Causal inference plays a vital role in diverse domains like epidemiology, healthcare, and economics. De-confounding and counterfactual prediction in observational data has emerged as a prominent concern in causal inference research. While existing models tackle observed confounders, the presence of unobserved confounders remains a significant challenge, distorting causal inference and impacting counterfactual outcome accuracy. To address this, we propose a novel variational learning model of unobserved confounders for counterfactual inference (VLUCI), which generates the posterior distribution of unobserved confounders. VLUCI relaxes the unconfoundedness assumption often overlooked by most causal inference methods. By disentangling observed and unobserved confounders, VLUCI constructs a doubly variational inference model to approximate the distribution of unobserved confounders, which are used for inferring more accurate counterfactual outcomes. Extensive experiments on synthetic and semi-synthetic datasets demonstrate VLUCI's superior performance in inferring unobserved confounders. It is compatible with state-of-the-art counterfactual inference models, significantly improving inference accuracy at both group and individual levels. Additionally, VLUCI provides confidence intervals for counterfactual outcomes, aiding decision-making in risk-sensitive domains. We further clarify the considerations when applying VLUCI to cases where unobserved confounders don't strictly conform to our model assumptions using the public IHDP dataset as an example, highlighting the practical advantages of VLUCI.