Deep Disentangled Representation Network for Treatment Effect Estimation

TL;DR

This paper introduces a novel disentangled representation approach for treatment effect estimation that combines a mixture of experts, multi-head attention, and orthogonal regularization to improve causal inference from observational data.

Contribution

It proposes a new algorithm that effectively decomposes covariates and reduces bias, outperforming existing methods in treatment effect estimation.

Findings

Outperforms state-of-the-art methods on semi-synthetic datasets

Demonstrates effectiveness on real-world datasets

Successfully disentangles covariate factors for causal inference

Abstract

Estimating individual-level treatment effect from observational data is a fundamental problem in causal inference and has attracted increasing attention in the fields of education, healthcare, and public policy.In this work, we concentrate on the study of disentangled representation methods that have shown promising outcomes by decomposing observed covariates into instrumental, confounding, and adjustment factors. However, most of the previous work has primarily revolved around generative models or hard decomposition methods for covariates, which often struggle to guarantee the attainment of precisely disentangled factors. In order to effectively model different causal relationships, we propose a novel treatment effect estimation algorithm that incorporates a mixture of experts with multi-head attention and a linear orthogonal regularizer to softly decompose the pre-treatment variables, and simultaneously eliminates selection bias via importance sampling re-weighting techniques. We conduct extensive experiments on both public semi-synthetic and real-world production datasets. The experimental results clearly demonstrate that our algorithm outperforms the state-of-the-art methods focused on individual treatment effects.