# Targeted VAE: Variational and Targeted Learning for Causal Inference

arXiv: https://arxiv.org/abs/2009.13472 · HTML: https://mlsanity.com/paper/2009.13472/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 28 Sep 2020.

## Built on by (most cited first, 7 of 7)

- [Deep Multi-Modal Structural Equations For Causal Effect Estimation With Unstructured Proxies](https://mlsanity.com/paper/2203.09672/) (2022)
- [Causal Inference from Text: Unveiling Interactions between Variables](https://mlsanity.com/paper/2311.05286/) (2023)
- [On the Sparse DAG Structure Learning Based on Adaptive Lasso](https://mlsanity.com/paper/2209.02946/) (2022)
- [On the Effects of Irrelevant Variables in Treatment Effect Estimation with Deep Disentanglement](https://mlsanity.com/paper/2407.20003/) (2024)
- [ACTIVA: Amortized Causal Effect Estimation via Transformer-based Variational Autoencoder](https://mlsanity.com/paper/2503.01290/) (2025)
- [Disentangled Graph Autoencoder for Treatment Effect Estimation](https://mlsanity.com/paper/2412.14497/) (2024)
- [MOCA: A Transformer-based Modular Causal Inference Framework with One-way Cross-attention and Cutting Feedback](https://mlsanity.com/paper/2604.23107/) (2026)

## Abstract (arXiv)

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 (textiti.e., differences between the treated and untreated groups), and an absence of counterfactual data (textiti.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.