DRTCI: Learning Disentangled Representations for Temporal Causal Inference

TL;DR

This paper introduces DRTCI, a method for learning disentangled representations in temporal data to improve causal inference in medical treatment evaluation, addressing the challenge of unknown confounders.

Contribution

It proposes a novel approach to disentangle covariate representations over time, enhancing the accuracy of causal inference in dynamic treatment settings.

Findings

Demonstrates improved causal effect estimation in temporal data

Effectively isolates confounding factors from predictive factors

Outperforms existing methods in simulation studies

Abstract

Medical professionals evaluating alternative treatment plans for a patient often encounter time varying confounders, or covariates that affect both the future treatment assignment and the patient outcome. The recently proposed Counterfactual Recurrent Network (CRN) accounts for time varying confounders by using adversarial training to balance recurrent historical representations of patient data. However, this work assumes that all time varying covariates are confounding and thus attempts to balance the full state representation. Given that the actual subset of covariates that may in fact be confounding is in general unknown, recent work on counterfactual evaluation in the static, non-temporal setting has suggested that disentangling the covariate representation into separate factors, where each either influence treatment selection, patient outcome or both can help isolate selection bias and restrict balancing efforts to factors that influence outcome, allowing the remaining factors which predict treatment without needlessly being balanced.