mlsanityData to 7 Oct 2026

On the Quantitative Analysis of Decoder-Based Generative Models

arXiv 1611.04273 ↗First posted 14 Nov 2016cs.LGcited by 40

Not on the map. Why?

Abstract (arXiv)

The past several years have seen remarkable progress in generative models which produce convincing samples of images and other modalities. A shared component of many powerful generative models is a decoder network, a parametric deep neural net that defines a generative distribution. Examples include variational autoencoders, generative adversarial networks, and generative moment matching networks. Unfortunately, it can be difficult to quantify the performance of these models because of the intractability of log-likelihood estimation, and inspecting samples can be misleading. We propose to use Annealed Importance Sampling for evaluating log-likelihoods for decoder-based models and validate its accuracy using bidirectional Monte Carlo. The evaluation code is provided at https://github.com/tonywu95/eval_gen. Using this technique, we analyze the performance of decoder-based models, the effectiveness of existing log-likelihood estimators, the degree of overfitting, and the degree to which these models miss important modes of the data distribution.

Key facts. Cited by 40 papers in the map (data through 7 Oct 2026). First posted on arXiv 14 Nov 2016.

Builds on

No reference list yet.

Built on by

Cited by year

Citing papers per year: 2021: 17; 2022: 8; 2023: 10; 2024: 5.

40 papers.

Sort:

By area:

  1. 2021How Faithful is your Synthetic Data? Sample-level Metrics for Evaluating and Auditing Generative Modelscited by 193
  2. 2021Annealed Flow Transport Monte Carlocited by 90
  3. 2021On Memorization in Probabilistic Deep Generative Modelscited by 60
  4. 2024Machine Unlearning for Image-to-Image Generative Modelscited by 47
  5. 2021MCMC Variational Inference via Uncorrected Hamiltonian Annealingcited by 39
  6. 2021Monte Carlo Variational Auto-Encoderscited by 36
  7. 2021Differentiable Annealed Importance Sampling and the Perils of Gradient Noisecited by 36
  8. 2023Feature Likelihood Divergence: Evaluating the Generalization of Generative Models Using Samplescited by 25
  9. 2021Generative Modeling for Multi-task Visual Learningcited by 14
  10. 2023Boosting Offline Reinforcement Learning for Autonomous Driving with Hierarchical Latent Skillscited by 13
  11. 2021q-Paths: Generalizing the Geometric Annealing Path using Power Meanscited by 12
  12. 2022ProbNeRF: Uncertainty-Aware Inference of 3D Shapes from 2D Imagescited by 12
28 more
  1. 2022Deterministic Langevin Monte Carlo with Normalizing Flows for Bayesian Inferencecited by 10
  2. 2022Transport Score Climbing: Variational Inference Using Forward KL and Adaptive Neural Transportcited by 8
  3. 2023Uncertainty-Aware Decision Transformer for Stochastic Driving Environmentscited by 8
  4. 2023Compositional diversity in visual concept learningcited by 7
  5. 2023Adaptive Annealed Importance Sampling with Constant Rate Progresscited by 7
  6. 2021Interpreting intermediate convolutional layers of generative CNNs trained on waveformscited by 6
  7. 2022Cooperation in the Latent Space: The Benefits of Adding Mixture Components in Variational Autoencoderscited by 6
  8. 2023Matching aggregate posteriors in the variational autoencodercited by 6
  9. 2024SeriesGAN: Time Series Generation via Adversarial and Autoregressive Learningcited by 6
  10. 2021Bootstrap Your Flowcited by 5
  11. 2023Fast-NTK: Parameter-Efficient Unlearning for Large-Scale Modelscited by 5
  12. 2021Generative Adversarial Networks and Adversarial Autoencoders: Tutorial and Surveycited by 4
  13. 2022Variational Inference with Locally Enhanced Bounds for Hierarchical Modelscited by 4
  14. 2023Diffusing Gaussian Mixtures for Generating Categorical Datacited by 4
  15. 2021Variational Inference with Holder Boundscited by 3
  16. 2022Estimators of Entropy and Information via Inference in Probabilistic Modelscited by 3
  17. 2024Variational Learning of Gaussian Process Latent Variable Models through Stochastic Gradient Annealed Importance Samplingcited by 3
  18. 2023Steering Language Generation: Harnessing Contrastive Expert Guidance and Negative Prompting for Coherent and Diverse Synthetic Data Generationcited by 2
  19. 2021NEO: Non Equilibrium Sampling on the Orbit of a Deterministic Transformcited by 1
  20. 2021Analyzing High-Resolution Clouds and Convection using Multi-Channel VAEscited by 1
  21. 2022Optimization of Annealed Importance Sampling Hyperparameterscited by 1
  22. 2023A Note on Generalization in Variational Autoencoders: How Effective Is Synthetic Data & Overparameterization?cited by 1
  23. 2024Understanding and mitigating difficulties in posterior predictive evaluationcited by 1
  24. 2021Deep Probabilistic Graphical Modelingcited by 0
  25. 2021Expectation Programming: Adapting Probabilistic Programming Systems to Estimate Expectations Efficientlycited by 0
  26. 2021A Simple Generative Networkcited by 0
  27. 2022Evaluation of Categorical Generative Models -- Bridging the Gap Between Real and Synthetic Datacited by 0
  28. 2024Intriguing Properties of Modern GANscited by 0