Targeted VAE: Variational and Targeted Learning for Causal Inference
This work addresses causal inference problems for applications like medical treatments and policy-making, but it appears incremental as it builds on existing methods like variational autoencoders and targeted learning.
The paper tackles the challenges of treatment assignment heterogeneity and lack of counterfactual data in causal inference with observational data by combining structured inference and targeted learning, resulting in competitive and state-of-the-art performance on benchmark datasets.
Undertaking causal inference with observational data is incredibly useful across a wide range of tasks including the development of medical treatments, advertisements and marketing, and policy making. There are two significant challenges associated with undertaking causal inference using observational data: treatment assignment heterogeneity (\textit{i.e.}, differences between the treated and untreated groups), and an absence of counterfactual data (\textit{i.e.}, not knowing what would have happened if an individual who did get treatment, were instead to have not been treated). We address these two challenges by combining structured inference and targeted learning. In terms of structure, we factorize the joint distribution into risk, confounding, instrumental, and miscellaneous factors, and in terms of targeted learning, we apply a regularizer derived from the influence curve in order to reduce residual bias. An ablation study is undertaken, and an evaluation on benchmark datasets demonstrates that TVAE has competitive and state of the art performance.