mlsanityData to 7 Oct 2026

A Critical Look at the Consistency of Causal Estimation With Deep Latent Variable Models

arXiv 2102.06648 ↗First posted 12 Feb 2021cs.LGcited by 29

Not on the map. Why?

Abstract (arXiv)

Using deep latent variable models in causal inference has attracted considerable interest recently, but an essential open question is their ability to yield consistent causal estimates. While they have demonstrated promising results and theory exists on some simple model formulations, we also know that causal effects are not even identifiable in general with latent variables. We investigate this gap between theory and empirical results with analytical considerations and extensive experiments under multiple synthetic and real-world data sets, using the causal effect variational autoencoder (CEVAE) as a case study. While CEVAE seems to work reliably under some simple scenarios, it does not estimate the causal effect correctly with a misspecified latent variable or a complex data distribution, as opposed to its original motivation. Hence, our results show that more attention should be paid to ensuring the correctness of causal estimates with deep latent variable models.

Key facts. Cited by 29 papers in the map (data through 7 Oct 2026). Builds on 14 of its 29 references that are in the map. First posted on arXiv 12 Feb 2021.

Builds on

14 of 29 references.

References · 14

  1. 2013Auto-Encoding Variational Bayescited by 3,980
  2. 2018Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representationscited by 929
  3. 2017Causal Effect Inference with Deep Latent-Variable Modelscited by 408
  4. 2019Adapting Neural Networks for the Estimation of Treatment Effectscited by 285
  5. 2020Deep Structural Causal Models for Tractable Counterfactual Inferencecited by 210
  6. 2016Identifying Causal Effects With Proxy Variables of an Unmeasured Confoundercited by 138
  7. 2019Lagging Inference Networks and Posterior Collapse in Variational Autoencoderscited by 124
  8. 2019Preventing Posterior Collapse with delta-VAEscited by 86
  9. 2020Identifying Causal-Effect Inference Failure with Uncertainty-Aware Modelscited by 57
  10. 2019The Usual Suspects? Reassessing Blame for VAE Posterior Collapsecited by 52
  11. 2018Fairness Through Causal Awareness: Learning Latent-Variable Models for Biased Datacited by 39
  12. 2019Counterfactual Reasoning for Fair Clinical Risk Predictioncited by 28
2 more
  1. 2018Linked Causal Variational Autoencoder for Inferring Paired Spillover Effectscited by 16
  2. 2020MissDeepCausal: Causal Inference from Incomplete Data Using Deep Latent Variable Modelscited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 2; 2022: 7; 2023: 7; 2024: 7; 2025: 4; 2026: 2.

29 papers.

Sort:

By area:

  1. 2023Measuring axiomatic soundness of counterfactual image modelscited by 34
  2. 2022Conceptualizing Treatment Leakage in Text-based Causal Inferencecited by 14
  3. 2022Adapting to Latent Subgroup Shifts via Concepts and Proxiescited by 12
  4. 2025DeCaFlow: A deconfounding causal generative modelcited by 8
  5. 2023Estimating Causal Effects using a Multi-task Deep Ensemblecited by 7
  6. 2023Delphic Offline Reinforcement Learning under Nonidentifiable Hidden Confoundingcited by 7
  7. 2024Continuous Treatment Effect Estimation Using Gradient Interpolation and Kernel Smoothingcited by 7
  8. 2024Conformal Convolution and Monte Carlo Meta-learners for Predictive Inference of Individual Treatment Effectscited by 7
  9. 2022Causal Inference with Conditional Instruments using Deep Generative Modelscited by 6
  10. 2024Proxy Methods for Domain Adaptationcited by 6
  11. 2024Disentangled Representation Learning for Causal Inference with Instrumentscited by 6
  12. 2022Time Series Causal Link Estimation under Hidden Confounding using Knockoff Interventionscited by 5
17 more
  1. 2023Kernel Single Proxy Control for Deterministic Confoundingcited by 5
  2. 2024Two-way Deconfounder for Off-policy Evaluation in Causal Reinforcement Learningcited by 5
  3. 2025Transfer Learning in Latent Contextual Bandits with Covariate Shift Through Causal Transportabilitycited by 4
  4. 2021SBI: A Simulation-Based Test of Identifiability for Bayesian Causal Inferencecited by 3
  5. 2022Causal Inference from Small High-dimensional Datasetscited by 3
  6. 2022Causal Deep Reinforcement Learning Using Observational Datacited by 3
  7. 2023High Dimensional Causal Inference with Variational Backdoor Adjustmentcited by 3
  8. 2024PairNet: Training with Observed Pairs to Estimate Individual Treatment Effectcited by 3
  9. 2022Confounder Balancing for Instrumental Variable Regression with Latent Variablecited by 1
  10. 2023Active and Passive Causal Inference Learningcited by 1
  11. 2021Multi-treatment Effect Estimation from Biomedical Datacited by 0
  12. 2023Identifiable causal inference with noisy treatment and no side informationcited by 0
  13. 2024Estimating Individual Dose-Response Curves under Unobserved Confounders from Observational Datacited by 0
  14. 2025Coupling Generative Modeling and an Autoencoder with the Causal Bridgecited by 0
  15. 2025PIPCFR: Pseudo-outcome Imputation with Post-treatment Variables for Individual Treatment Effect Estimationcited by 0
  16. 2026Uncertainty-Guided LLM Semantic Augmentation for Heterogeneous Treatment Effect Estimationcited by 0
  17. 2026Proximal Balancing for Causal Effect Estimation under Unmeasured Confoundingcited by 0