mlsanityData to 14 Oct 2026

Matching aggregate posteriors in the variational autoencoder

arXiv 2311.07693 ↗First posted 13 Nov 2023cs.LGcited by 6

Not on the map. Why?

Abstract (arXiv)

The variational autoencoder (VAE) is a well-studied, deep, latent-variable model (DLVM) that efficiently optimizes the variational lower bound of the log marginal data likelihood and has a strong theoretical foundation. However, the VAE's known failure to match the aggregate posterior often results in pockets/holes in the latent distribution (i.e., a failure to match the prior) and/or posterior collapse, which is associated with a loss of information in the latent space. This paper addresses these shortcomings in VAEs by reformulating the objective function associated with VAEs in order to match the aggregate/marginal posterior distribution to the prior. We use kernel density estimate (KDE) to model the aggregate posterior in high dimensions. The proposed method is named the aggregate variational autoencoder (AVAE) and is built on the theoretical framework of the VAE. Empirical evaluation of the proposed method on multiple benchmark data sets demonstrates the effectiveness of the AVAE relative to state-of-the-art (SOTA) methods.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 59 of its 94 references that are in the map. First posted on arXiv 13 Nov 2023.

Builds on

59 of 94 references.

References · 59

  1. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  2. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 9,122
  3. 2017Neural Discrete Representation Learningcited by 4,712
  4. 2014Deep Learning Face Attributes in the Wildcited by 3,693
  5. 2016Improved Techniques for Training GANscited by 2,970
  6. 2017Improved Training of Wasserstein GANscited by 2,235
  7. 2015Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networkscited by 2,054
  8. 2016Density estimation using Real NVPcited by 2,042
  9. 2021Alias-Free Generative Adversarial Networkscited by 1,184
  10. 2016Least Squares Generative Adversarial Networkscited by 1,001
  11. 2018Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representationscited by 941
  12. 2018Which Training Methods for GANs do actually Converge?cited by 679
38 more
  1. 2018Disentangling by Factorisingcited by 616
  2. 2015Generating Sentences from a Continuous Spacecited by 612
  3. 2016f-GAN: Training Generative Neural Samplers using Variational Divergence Minimizationcited by 506
  4. 2017Towards Principled Methods for Training Generative Adversarial Networkscited by 464
  5. 2015Importance Weighted Autoencoderscited by 456
  6. 2018Assessing Generative Models via Precision and Recallcited by 406
  7. 2017Wasserstein Auto-Encoderscited by 373
  8. 2019Deep Learning for Anomaly Detection: A Surveycited by 358
  9. 2016Adversarial Feature Learningcited by 350
  10. 2016Generative Visual Manipulation on the Natural Image Manifoldcited by 348
  11. 2015MADE: Masked Autoencoder for Distribution Estimationcited by 307
  12. 2015A note on the evaluation of generative modelscited by 292
  13. 2016NIPS 2016 Tutorial: Generative Adversarial Networkscited by 285
  14. 2017Are GANs Created Equal? A Large-Scale Studycited by 267
  15. 2015Generative Moment Matching Networkscited by 257
  16. 2017MMD GAN: Towards Deeper Understanding of Moment Matching Networkcited by 254
  17. 2017VAE with a VampPriorcited by 248
  18. 2017Variational Inference of Disentangled Latent Concepts from Unlabeled Observationscited by 241
  19. 2016Adversarially Learned Inferencecited by 216
  20. 2018Adversarially Learned One-Class Classifier for Novelty Detectioncited by 202
  21. 2019Diagnosing and Enhancing VAE Modelscited by 193
  22. 2019Cyclical Annealing Schedule: A Simple Approach to Mitigating KL Vanishingcited by 191
  23. 2018Isolating Sources of Disentanglement in Variational Autoencoderscited by 167
  24. 2017Generalization and Equilibrium in Generative Adversarial Nets (GANs)cited by 166
  25. 2018Recent Advances in Autoencoder-Based Representation Learningcited by 133
  26. 2019From Variational to Deterministic Autoencoderscited by 133
  27. 2019Don't Blame the ELBO! A Linear VAE Perspective on Posterior Collapsecited by 121
  28. 2017Stabilizing Training of Generative Adversarial Networks through Regularizationcited by 118
  29. 2016Deep Structured Energy Based Models for Anomaly Detectioncited by 114
  30. 2021Generating Diverse Structure for Image Inpainting With Hierarchical VQ-VAEcited by 106
  31. 2018Generative Probabilistic Novelty Detection with Adversarial Autoencoderscited by 99
  32. 2018Variational Autoencoders Pursue PCA Directions (by Accident)cited by 85
  33. 2020Measuring Disentanglement: A Review of Metricscited by 70
  34. 2018Resampled Priors for Variational Autoencoderscited by 57
  35. 2020Simple and Effective VAE Training with Calibrated Decoderscited by 57
  36. 2017Variational Approaches for Auto-Encoding Generative Adversarial Networkscited by 52
  37. 2019The Usual Suspects? Reassessing Blame for VAE Posterior Collapsecited by 52
  38. 2016On the Quantitative Analysis of Decoder-Based Generative Modelscited by 40

50 most cited of 59 All 59 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 2; 2025: 3; 2026: 1.

6 papers.

  1. 2025ARD-VAE: A Statistical Formulation to Find the Relevant Latent Dimensions of Variational Autoencoderscited by 6
  2. 2024On Kernel-based Variational Autoencodercited by 2
  3. 2024Analyzing Multimodal Integration in the Variational Autoencoder from an Information-Theoretic Perspectivecited by 1
  4. 2025Disentanglement Analysis in Deep Latent Variable Models Matching Aggregate Posterior Distributionscited by 1
  5. 2025AdaSemSeg: An Adaptive Few-shot Semantic Segmentation of Seismic Faciescited by 0
  6. 2026Yield Curves Dynamics Using Variational Autoencoders Under No-arbitragecited by 0