Non parametric estimation of causal populations in a counterfactual scenario

TL;DR

This paper introduces a novel method using a Causal Auto-Encoder to estimate the distribution of causal populations in scenarios where treatment effects are unobservable due to confounding, reformulating the problem as a missing data model.

Contribution

It proposes a new approach with a Causal Auto-Encoder that incorporates treatment and outcome information to estimate hidden causal population distributions.

Findings

Effective estimation of causal populations demonstrated

Reconstruction accuracy improved with the proposed model

Framework applicable to various causal inference scenarios

Abstract

In causality, estimating the effect of a treatment without confounding inference remains a major issue because requires to assess the outcome in both case with and without treatment. Not being able to observe simultaneously both of them, the estimation of potential outcome remains a challenging task. We propose an innovative approach where the problem is reformulated as a missing data model. The aim is to estimate the hidden distribution of \emph{causal populations}, defined as a function of treatment and outcome. A Causal Auto-Encoder (CAE), enhanced by a prior dependent on treatment and outcome information, assimilates the latent space to the probability distribution of the target populations. The features are reconstructed after being reduced to a latent space and constrained by a mask introduced in the intermediate layer of the network, containing treatment and outcome information.