🤖 AI Summary
Estimating causal effects from non-randomized observational data—such as electronic health records—is challenging due to confounding bias. To address this, we propose a unified causal inference framework integrating double robust estimation, variational autoencoders (VAEs), and information-theoretic generative adversarial networks (Info-GANs). Our method employs latent variable modeling to disentangle confounders and incorporates mutual information constraints to enhance the reliability of counterfactual generation. The double robust structure guarantees unbiased estimation of individualized treatment effects (ITE) even if either the propensity score or outcome regression model is misspecified. Extensive experiments on synthetic and real-world benchmarks—including IHDP, Twin, and NSW—demonstrate that our approach consistently outperforms state-of-the-art generative and non-generative baselines, achieving superior accuracy and robustness in ITE estimation.
📝 Abstract
Determining causal effects of interventions onto outcomes from real-world, observational (non-randomized) data, e.g., treatment repurposing using electronic health records, is challenging due to underlying bias. Causal deep learning has improved over traditional techniques for estimating individualized treatment effects (ITE). We present the Doubly Robust Variational Information-theoretic Deep Adversarial Learning (DR-VIDAL), a novel generative framework that combines two joint models of treatment and outcome, ensuring an unbiased ITE estimation even when one of the two is misspecified. DR-VIDAL integrates: (i) a variational autoencoder (VAE) to factorize confounders into latent variables according to causal assumptions; (ii) an information-theoretic generative adversarial network (Info-GAN) to generate counterfactuals; (iii) a doubly robust block incorporating treatment propensities for outcome predictions. On synthetic and real-world datasets (Infant Health and Development Program, Twin Birth Registry, and National Supported Work Program), DR-VIDAL achieves better performance than other non-generative and generative methods. In conclusion, DR-VIDAL uniquely fuses causal assumptions, VAE, Info-GAN, and doubly robustness into a comprehensive, per- formant framework. Code is available at: https://github.com/Shantanu48114860/DR-VIDAL-AMIA-22 under MIT license.
