mlsanityData to 14 Oct 2026

Reinterpreting Importance-Weighted Autoencoders

arXiv 1704.02916 ↗First posted 10 Apr 2017stat.MLcited by 37

Not on the map. Why?

Abstract (arXiv)

The standard interpretation of importance-weighted autoencoders is that they maximize a tighter lower bound on the marginal likelihood than the standard evidence lower bound. We give an alternate interpretation of this procedure: that it optimizes the standard variational lower bound, but using a more complex distribution. We formally derive this result, present a tighter lower bound, and visualize the implicit importance-weighted distribution.

Key facts. Cited by 37 papers in the map (data through 14 Oct 2026). First posted on arXiv 10 Apr 2017.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2019: 1; 2020: 0; 2021: 11; 2022: 11; 2023: 4; 2024: 2; 2025: 3; 2026: 5.

37 papers.

Sort:

By area:

  1. 2019Variational Mixture-of-Experts Autoencoders for Multi-Modal Deep Generative Modelscited by 178
  2. 2025Reasoning to Learn from Latent Thoughtscited by 45
  3. 2022Variational methods for simulation-based inferencecited by 43
  4. 2021MCMC Variational Inference via Uncorrected Hamiltonian Annealingcited by 39
  5. 2021Decomposed Mutual Information Estimation for Contrastive Representation Learningcited by 25
  6. 2021Improving Lossless Compression Rates via Monte Carlo Bits-Back Codingcited by 19
  7. 2021Surrogate Likelihoods for Variational Annealed Importance Samplingcited by 14
  8. 2021A likelihood approach to nonparametric estimation of a singular distribution using deep generative modelscited by 12
  9. 2022Recursive Monte Carlo and Variational Inference with Auxiliary Variablescited by 8
  10. 2022Variational Sparse Coding with Learned Thresholdingcited by 6
  11. 2021Scalable Multi-Task Gaussian Processes with Neural Embedding of Coregionalizationcited by 5
  12. 2022Multi-Sample Training for Neural Image Compressioncited by 5
25 more
  1. 2022Variational Inference with Locally Enhanced Bounds for Hierarchical Modelscited by 4
  2. 2022Black-box Coreset Variational Inferencecited by 4
  3. 2021An Easy to Interpret Diagnostic for Approximate Inference: Symmetric Divergence Over Simulationscited by 3
  4. 2021Deep Generative Pattern-Set Mixture Models for Nonignorable Missingnesscited by 3
  5. 2023U-Statistics for Importance-Weighted Variational Inferencecited by 3
  6. 2026Self-Supervised Bootstrapping of Action-Predictive Embodied Reasoningcited by 3
  7. 2022Deeply-Learned Generalized Linear Models with Missing Datacited by 2
  8. 2023Unscented Autoencodercited by 2
  9. 2023Reparameterized Variational Rejection Samplingcited by 2
  10. 2024Improving Variational Autoencoder Estimation from Incomplete Data with Mixture Variational Familiescited by 2
  11. 2024Differentiable Annealed Importance Sampling Minimizes The Symmetrized Kullback-Leibler Divergence Between Initial and Target Distributioncited by 2
  12. 2021Unsupervised Imputation of Non-ignorably Missing Data Using Importance-Weighted Autoencoderscited by 1
  13. 2021NEO: Non Equilibrium Sampling on the Orbit of a Deterministic Transformcited by 1
  14. 2022Latent Variable Modelling Using Variational Autoencoders: A surveycited by 1
  15. 2022Decomposed Mutual Information Optimization for Generalized Context in Meta-Reinforcement Learningcited by 1
  16. 2023Massively Parallel Reweighted Wake-Sleepcited by 1
  17. 2021Deep Probabilistic Graphical Modelingcited by 0
  18. 2022Tighter Variational Bounds are Not Necessarily Better. A Research Report on Implementation, Ablation Study, and Extensionscited by 0
  19. 2022Generalizing Multimodal Variational Methods to Setscited by 0
  20. 2025Massively Parallel Expectation Maximization For Approximate Posteriorscited by 0
  21. 2025Misspecification-robust amortised simulation-based inference using variational methodscited by 0
  22. 2026Bures-Wasserstein Importance-Weighted Evidence Lower Bound: Exposition and Applicationscited by 0
  23. 2026MDP Planning as Policy Inferencecited by 0
  24. 2026Structured SIR: Efficient and Expressive Importance-Weighted Inference for High-Dimensional Image Registrationcited by 0
  25. 2026Universal Bounds for Self-Normalized Kullback-Leibler Estimationcited by 0