Conditional Instrumental Variable Regression with Representation Learning for Causal Inference

TL;DR

This paper introduces CBRL.CIV, a novel non-linear conditional instrumental variable regression method with representation learning, to estimate causal effects from observational data without relying on linearity or strict unconfoundedness assumptions.

Contribution

It proposes a non-linear CIV regression approach with confounding balancing representation learning, relaxing key assumptions of standard IV methods for better causal inference.

Findings

CBRL.CIV outperforms existing IV-based estimators on synthetic data.

It demonstrates superior performance in non-linear causal inference scenarios.

The method effectively balances observed and unobserved confounders.

Abstract

This paper studies the challenging problem of estimating causal effects from observational data, in the presence of unobserved confounders. The two-stage least square (TSLS) method and its variants with a standard instrumental variable (IV) are commonly used to eliminate confounding bias, including the bias caused by unobserved confounders, but they rely on the linearity assumption. Besides, the strict condition of unconfounded instruments posed on a standard IV is too strong to be practical. To address these challenging and practical problems of the standard IV method (linearity assumption and the strict condition), in this paper, we use a conditional IV (CIV) to relax the unconfounded instrument condition of standard IV and propose a non-linear CIV regression with Confounding Balancing Representation Learning, CBRL.CIV, for jointly eliminating the confounding bias from unobserved confounders and balancing the observed confounders, without the linearity assumption. We theoretically demonstrate the soundness of CBRL.CIV. Extensive experiments on synthetic and two real-world datasets show the competitive performance of CBRL.CIV against state-of-the-art IV-based estimators and superiority in dealing with the non-linear situation.