Deep Treatment-Adaptive Network for Causal Inference

TL;DR

This paper introduces a deep treatment-adaptive network designed to improve causal inference by effectively handling both pre-treatment and post-treatment covariates, addressing biases caused by treatment effects on covariates.

Contribution

The proposed method adaptively learns representations that account for post-treatment covariates, overcoming limitations of existing methods that assume all covariates are unaffected by treatment.

Findings

Demonstrates improved treatment effect estimation accuracy.

Effectively handles post-treatment covariates in observational data.

Outperforms traditional representation-based methods.

Abstract

Causal inference is capable of estimating the treatment effect (i.e., the causal effect of treatment on the outcome) to benefit the decision making in various domains. One fundamental challenge in this research is that the treatment assignment bias in observational data. To increase the validity of observational studies on causal inference, representation based methods as the state-of-the-art have demonstrated the superior performance of treatment effect estimation. Most representation based methods assume all observed covariates are pre-treatment (i.e., not affected by the treatment), and learn a balanced representation from these observed covariates for estimating treatment effect. Unfortunately, this assumption is often too strict a requirement in practice, as some covariates are changed by doing an intervention on treatment (i.e., post-treatment). By contrast, the balanced representation learned from unchanged covariates thus biases the treatment effect estimation.