Research map / Image generation / GANs
Bayesian GAN
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Bayesian marginalization of generator and discriminator weights via SGHMC (vs. standard DCGAN) | DCGAN on semi-supervised benchmarks · MNIST N=20/50/100/200, CIFAR-10, SVHN, CelebA | test error / misclassifications | Helped |
| Ensemble of 10 DCGANs vs single BayesGAN | MNIST N=100 · MNIST semi-supervised | misclassifications: 102 vs 98 | Hurt |
| Using Adam for first few thousand SGHMC iterations (burn-in) before momentum SGHMC | BayesGAN training · all experiments | qualitative | No clear effect |
| Omitting standard GAN stabilization tricks (minibatch discrimination, feature normalization, Gaussian noise) by using Bayesian formulation | BayesGAN vs DCGAN · MNIST sample generation | qualitative sample quality | Helped |
| Gaussian N(0,10I) prior on generator and discriminator weights | BayesGAN · all experiments | qualitative robustness | Helped |
| Bayesian model averaging over posterior samples of discriminator weights | BayesGAN semi-supervised classification · MNIST, CIFAR-10, SVHN, CelebA | test error | Helped |
Closest papers by meaning
- ProbGAN: Towards Probabilistic GAN with Theoretical Guarantees (ICLR 2019 · similarity 0.92)
- Variational Annealing of GANs: A Langevin Perspective (ICML 2019 · similarity 0.91)
- Metropolis-Hastings Generative Adversarial Networks (ICML 2019 · similarity 0.91)
- MCGAN: Enhancing GAN Training with Regression-Based Generator Loss (AAAI 2025 · similarity 0.90)
- Bayesian GANs (NeurIPS 2017 · similarity 0.89)
- Direct Discriminative Optimization: Your Likelihood-Based Visual Generative Model is Secretly a GAN Discriminator (ICML 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.