arXiv 30 Sep 2021 · Statistics — Machine Learning · 1 citations (OpenAlex)
arXiv:2109.15062 · PDF · DOI · OpenAlex · Extracted main text
As an important problem in causal inference, we discuss the estimation of treatment effects (TEs). Representing the confounder as a latent variable, we propose Intact-VAE, a new variant of variational autoencoder (VAE), motivated by the prognostic score that is sufficient for identifying TEs. Our VAE also naturally gives representations balanced for treatment groups, using its prior. Experiments on (semi-)synthetic datasets show state-of-the-art performance under diverse settings, including unobserved confounding. Based on the identifiability of our model, we prove identification of TEs under unconfoundedness, and also discuss (possible) extensions to harder settings.
appendix boundary found by appendix_command · 55% of the source is main text. Read the extracted text to check this.
The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.
| Reference | Intensity | Mentions | Sections | Main text | |
|---|---|---|---|---|---|
| 1 | Christos Louizos, Uri Shalit, Joris M Mooij, David Sontag, Richard Z… (2017) Causal effect inference with deep latent-variable models | 1.000 | 6 | 3 | 100% |
| 2 | Uri Shalit, Fredrik D Johansson, and David Sontag (2017) Estimating individual treatment effect: generalization bounds and algorithms | 0.956 | 8 | 4 | 88% |
| 3 | Anonymous (2021) $$beta-intact-vae: Identifying and estimating causal effects under limited overlap | 0.941 | 6 | 3 | 83% |
| 4 | Victor Veitch, Yixin Wang, and David Blei (2019) Using embeddings to correct for unobserved confounding in networks | 0.874 | 7 | 2 | 100% |
| 5 | Ben B Hansen (2008) The prognostic analogue of the propensity score | 0.874 | 6 | 3 | 67% |
| 6 | Weijia Zhang, Lin Liu, and Jiuyong Li (2020) Treatment effect estimation with disentangled latent factors | 0.811 | 4 | 2 | 100% |
| 7 | Ilyes Khemakhem, Diederik Kingma, Ricardo Monti, and Aapo Hyvarinen (2020) Variational autoencoders and nonlinear ica: A unifying framework | 0.737 | 10 | 5 | 40% |
| 8 | Paul R Rosenbaum and Donald B Rubin (1983) The central role of the propensity score in observational studies for causal effects | 0.737 | 3 | 3 | 67% |
| 9 | Fredrik Johansson, Uri Shalit, and David Sontag (2016) Learning representations for counterfactual inference | 0.737 | 3 | 2 | 100% |
| 10 | Eric J Tchetgen Tchetgen, Andrew Ying, Yifan Cui, Xu Shi, and Wang M… (2020) An introduction to proximal causal learning | 0.737 | 3 | 2 | 100% |
Showing the top 10 of 81 scored citations.
arXiv econ.EM papers that cite this one, ranked by how heavily they lean on it.
| Citing paper | Intensity | Mentions | Sections | |
|---|---|---|---|---|
| 1 | $$-Intact-VAE: Identifying and Estimating Causal Effects under Limited Overlap | 0.644 | 2 | 2 |