mlsanityData to 14 Oct 2026

Bridging GANs and Bayesian Neural Networks via Partial Stochasticity

arXiv 2507.00651 ↗First posted 1 Jul 2025cs.LG cs.CV stat.MLcited by 0

Not on the map. Why?

Abstract (arXiv)

Generative Adversarial Networks (GANs) are popular and successful generative models. Despite their success, optimization is notoriously challenging. In this work, we explain the success and limitations of GANs by casting them as Bayesian neural networks with partial stochasticity. This interpretation allows us to establish conditions of universal approximation and to rewrite the adversarial-style optimization of several variants of GANs as the optimization of a proxy for the likelihood obtained by marginalizing out the stochastic variables. Following this interpretation, the need for regularization becomes apparent, and we propose to adopt strategies to smooth the loss landscape and methods to search for solutions with minimum description length, which are associated with flat minima and good generalization. Results obtained on a wide range of experiments indicate that these strategies lead to performance improvements and pave the way to a deeper understanding of GANs.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 31 of its 58 references that are in the map. First posted on arXiv 1 Jul 2025.

Builds on

31 of 58 references.

References · 31

  1. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 6,013
  2. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,448
  3. 2019Analyzing and Improving the Image Quality of StyleGANcited by 3,588
  4. 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,626
  5. 2017Improved Training of Wasserstein GANscited by 2,235
  6. 2015Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networkscited by 2,054
  7. 2020Sharpness-Aware Minimization for Efficiently Improving Generalizationcited by 1,417
  8. 2020Training Generative Adversarial Networks with Limited Datacited by 1,186
  9. 2021Alias-Free Generative Adversarial Networkscited by 1,184
  10. 2015Generative Moment Matching Networkscited by 257
  11. 2017MMD GAN: Towards Deeper Understanding of Moment Matching Networkcited by 254
  12. 2018The relativistic discriminator: a key element missing from standard GANcited by 238
19 more
  1. 2022Diffusion-GAN: Training GANs with Diffusioncited by 205
  2. 2022Fast Sampling of Diffusion Models via Operator Learningcited by 175
  3. 2015Training generative neural networks via Maximum Mean Discrepancy optimizationcited by 151
  4. 2017Stabilizing Training of Generative Adversarial Networks through Regularizationcited by 118
  5. 2017Adversarial Variational Bayes: Unifying Variational Autoencoders and Generative Adversarial Networkscited by 107
  6. 2019Learning with minibatch Wasserstein : asymptotic and gradient propertiescited by 87
  7. 2017Gradient descent GAN optimization is locally stablecited by 67
  8. 2022Verifying the Union of Manifolds Hypothesis for Image Datacited by 64
  9. 2022Do Bayesian Neural Networks Need To Be Fully Stochastic?cited by 63
  10. 2020All You Need is a Good Functional Prior for Bayesian Deep Learningcited by 61
  11. 2025The GAN is dead; long live the GAN! A Modern GAN Baselinecited by 59
  12. 2017Bayesian GANcited by 32
  13. 2022Diagnosing and Fixing Manifold Overfitting in Deep Generative Modelscited by 28
  14. 2022Improving Lipschitz-Constrained Neural Networks by Learning Activation Functionscited by 24
  15. 2017Wasserstein Divergence for GANscited by 22
  16. 2021Model Selection for Bayesian Autoencoderscited by 10
  17. 2023Balanced Training for Sparse GANscited by 8
  18. 2018Entropic GANs meet VAEs: A Statistical Approach to Compute Sample Likelihoods in GANscited by 4
  19. 2018Cycle-Consistent Adversarial Learning as Approximate Bayesian Inferencecited by 3

Built on by

Not cited yet.