mlsanityData to 14 Oct 2026

Failure Modes of Variational Autoencoders and Their Effects on Downstream Tasks

arXiv 2007.07124 ↗First posted 14 Jul 2020stat.ML cs.LGcited by 16

Not on the map. Why?

Abstract (arXiv)

Variational Auto-encoders (VAEs) are deep generative latent variable models that are widely used for a number of downstream tasks. While it has been demonstrated that VAE training can suffer from a number of pathologies, existing literature lacks characterizations of exactly when these pathologies occur and how they impact downstream task performance. In this paper, we concretely characterize conditions under which VAE training exhibits pathologies and connect these failure modes to undesirable effects on specific downstream tasks, such as learning compressed and disentangled representations, adversarial robustness, and semi-supervised learning.

Key facts. Cited by 16 papers in the map (data through 14 Oct 2026). First posted on arXiv 14 Jul 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 2; 2022: 2; 2023: 3; 2024: 3; 2025: 2; 2026: 4.

16 papers.

Sort:

  1. 2022Identifiability of deep generative models without auxiliary informationcited by 70
  2. 2023ARTxAI: Explainable Artificial Intelligence Curates Deep Representation Learning for Artistic Images using Fuzzy Techniquescited by 10
  3. 2023HyperSINDy: Deep Generative Modeling of Nonlinear Stochastic Governing Equationscited by 8
  4. 2021The Deep Generative Decoder: MAP estimation of representations improves modeling of single-cell RNA datacited by 2
  5. 2022Towards a Taxonomy for the Use of Synthetic Data in Advanced Analyticscited by 2
  6. 2024A Modular System for Enhanced Robustness of Multimedia Understanding Networks via Deep Parametric Estimationcited by 2
  7. 2025Data-driven Seasonal Climate Predictions via Variational Inference and Transformerscited by 1
  8. 2021Challenging the Semi-Supervised VAE Framework for Text Classificationcited by 0
  9. 2023Interpretable Sentence Representation with Variational Autoencoders and Attentioncited by 0
  10. 2024Towards Model-Agnostic Posterior Approximation for Fast and Accurate Variational Autoencoderscited by 0
  11. 2024Survey of Deep Learning and Physics-Based Approaches in Computational Wave Imagingcited by 0
  12. 2025PH-VAE: A Polynomial Hierarchical Variational Autoencoder Towards Disentangled Representation Learningcited by 0
4 more
  1. 2026Uncovering Physical Drivers of Dark Matter Halo Structures with Auxiliary-Variable-Guided Generative Modelscited by 0
  2. 2026eXact-Prior Variational Autoencoder (X-VAE): Learning Data-Adaptive Gaussian Mixture Priors for Latent Distributionscited by 0
  3. 2026Neural Operator-enabled Topology-informed Evolutionary Strategy for PDE-Constrained Optimizationcited by 0
  4. 2026Latent Similarity Gaussian Processes: A Theory-Grounded Approach to Personalized Suicide-Risk Forecasting for Clinical Decision-Supportcited by 0