mlsanityData to 7 Oct 2026

Towards Principled Causal Effect Estimation by Deep Identifiable Models

arXiv 2109.15062 ↗First posted 30 Sep 2021stat.ML cs.LG econ.EMcited by 4

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 33 of its 93 references that are in the map. First posted on arXiv 30 Sep 2021.

Builds on

33 of 93 references.

References · 33

  1. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
  2. 2013Sinkhorn Distances: Lightspeed Computation of Optimal Transportation Distancescited by 2,277
  3. 2019An Introduction to Variational Autoencoderscited by 997
  4. 2015Estimation and Inference of Heterogeneous Treatment Effects using Random Forestscited by 597
  5. 2014Semi-Supervised Learning with Deep Generative Modelscited by 494
  6. 2019Variational Autoencoders and Nonlinear ICA: A Unifying Frameworkcited by 494
  7. 2017Causal Effect Inference with Deep Latent-Variable Modelscited by 408
  8. 2016Tutorial on Variational Autoencoderscited by 404
  9. 2016Learning Representations for Counterfactual Inferencecited by 327
  10. 2019Adapting Neural Networks for the Estimation of Treatment Effectscited by 285
  11. 2018Nonlinear ICA Using Auxiliary Variables and Generalized Contrastive Learningcited by 278
  12. 2016Unsupervised Feature Extraction by Time-Contrastive Learning and Nonlinear ICAcited by 270
21 more
  1. 2017Bayesian Inference of Individualized Treatment Effects using Multi-task Gaussian Processescited by 156
  2. 2016Identifying Causal Effects With Proxy Variables of an Unmeasured Confoundercited by 138
  3. 2019Kernel Instrumental Variable Regressioncited by 118
  4. 2018Robustly Disentangled Causal Mechanisms: Validating Deep Representations for Interventional Robustnesscited by 108
  5. 2018The Blessings of Multiple Causescited by 103
  6. 2020On Linear Identifiability of Learned Representationscited by 103
  7. 2020ICE-BeeM: Identifiable Conditional Energy-Based Deep Models Based on Nonlinear ICAcited by 102
  8. 2020Disentanglement by Nonlinear ICA with General Incompressible-flow Networks (GIN)cited by 91
  9. 2020An Introduction to Proximal Causal Learningcited by 91
  10. 2020Minimax Estimation of Conditional Moment Modelscited by 83
  11. 2020Treatment effect estimation with disentangled latent factorscited by 63
  12. 2021Proximal Causal Learning with Kernels: Two-Stage Estimation and Moment Restrictioncited by 54
  13. 2018Interval Estimation of Individual-Level Causal Effects Under Unobserved Confoundingcited by 51
  14. 2019Using Embeddings to Correct for Unobserved Confounding in Networkscited by 31
  15. 2018Causal Inference with Noisy and Missing Covariates via Matrix Factorizationcited by 30
  16. 2020The Variational Method of Momentscited by 23
  17. 2020Causal Mosaic: Cause-Effect Inference via Nonlinear ICA and Ensemble Methodcited by 16
  18. 2020Increasing the efficiency of randomized trial estimates via linear adjustment for a prognostic scorecited by 14
  19. 2021Intact-VAE: Estimating Treatment Effects under Unobserved Confoundingcited by 10
  20. 2021Estimating Average Treatment Effects with Support Vector Machinescited by 3
  21. 2019Posterior Average Effectscited by 2

Built on by

Cited by year

Citing papers per year: 2021: 1; 2022: 1; 2023: 1; 2024: 0; 2025: 1.

4 papers.

  1. 2022Deep Multi-Modal Structural Equations For Causal Effect Estimation With Unstructured Proxiescited by 13
  2. 2021β-Intact-VAE: Identifying and Estimating Causal Effects under Limited Overlapcited by 8
  3. 2023A Survey of Methods, Challenges and Perspectives in Causalitycited by 6
  4. 2025Spatial Deconfounder: Interference-Aware Deconfounding for Spatial Causal Inferencecited by 3