Causal Effect Inference with Deep Latent-Variable Models

TLDR

Learning individual‑level causal effects from observational data is increasingly important, but requires careful handling of confounders, which may only be observed indirectly through noisy proxies. The study aims to jointly estimate a latent confounder space and causal effects using latent variable models. The authors employ a Variational Autoencoder that models the causal structure of inference with proxy variables. The proposed method outperforms existing approaches and achieves state‑of‑the‑art performance on individual treatment effect benchmarks.

Abstract

Learning individual-level causal effects from observational data, such as inferring the most effective medication for a specific patient, is a problem of growing importance for policy makers. The most important aspect of inferring causal effects from observational data is the handling of confounders, factors that affect both an intervention and its outcome. A carefully designed observational study attempts to measure all important confounders. However, even if one does not have direct access to all confounders, there may exist noisy and uncertain measurement of proxies for confounders. We build on recent advances in latent variable modeling to simultaneously estimate the unknown latent space summarizing the confounders and the causal effect. Our method is based on Variational Autoencoders (VAE) which follow the causal structure of inference with proxies. We show our method is significantly more robust than existing methods, and matches the state-of-the-art on previous benchmarks focused on individual treatment effects.