Targeted VAE: Variational and Targeted Learning for Causal Inference

TL;DR

This paper introduces Targeted VAE, a novel approach combining structured inference and targeted learning to improve causal inference from observational data, addressing treatment heterogeneity and counterfactual absence.

Contribution

It proposes a new VAE-based framework that factorizes the joint distribution and applies influence curve regularization for better causal effect estimation.

Findings

Achieves state-of-the-art performance on benchmark datasets.

Effectively reduces residual bias in causal inference.

Demonstrates robustness across different observational data scenarios.

Abstract

Undertaking causal inference with observational data is incredibly useful across a wide range of tasks including the development of medical treatments, advertisements and marketing, and policy making. There are two significant challenges associated with undertaking causal inference using observational data: treatment assignment heterogeneity (\textit{i.e.}, differences between the treated and untreated groups), and an absence of counterfactual data (\textit{i.e.}, not knowing what would have happened if an individual who did get treatment, were instead to have not been treated). We address these two challenges by combining structured inference and targeted learning. In terms of structure, we factorize the joint distribution into risk, confounding, instrumental, and miscellaneous factors, and in terms of targeted learning, we apply a regularizer derived from the influence curve in order to reduce residual bias. An ablation study is undertaken, and an evaluation on benchmark datasets demonstrates that TVAE has competitive and state of the art performance.