mlsanityData to 14 Oct 2026

On Convergence and Stability of GANs

arXiv 1705.07215 ↗First posted 19 May 2017cs.AI cs.CV cs.GTcited by 18

Not on the map. Why?

Abstract (arXiv)

We propose studying GAN training dynamics as regret minimization, which is in contrast to the popular view that there is consistent minimization of a divergence between real and generated distributions. We analyze the convergence of GAN training from this new point of view to understand why mode collapse happens. We hypothesize the existence of undesirable local equilibria in this non-convex game to be responsible for mode collapse. We observe that these local equilibria often exhibit sharp gradients of the discriminator function around some real data points. We demonstrate that these degenerate local equilibria can be avoided with a gradient penalty scheme called DRAGAN. We show that DRAGAN enables faster training, achieves improved stability with fewer mode collapses, and leads to generator networks with better modeling performance across a variety of architectures and objective functions.

Key facts. Cited by 18 papers in the map (data through 14 Oct 2026). First posted on arXiv 19 May 2017.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2018: 2; 2019: 0; 2020: 1; 2021: 2; 2022: 6; 2023: 2; 2024: 2; 2025: 2; 2026: 1.

18 papers.

Sort:

By area:

  1. 2018Variational Discriminator Bottleneck: Improving Imitation Learning, Inverse RL, and GANs by Constraining Information Flowcited by 105
  2. 2021Revisiting Discriminator in GAN Compression: A Generator-discriminator Cooperative Compression Schemecited by 23
  3. 2018MEGAN: Mixture of Experts of Generative Adversarial Networks for Multimodal Image Generationcited by 15
  4. 2020Lipschitzness Is All You Need To Tame Off-policy Generative Adversarial Imitation Learningcited by 15
  5. 2022Instability and Local Minima in GAN Training with Kernel Discriminatorscited by 7
  6. 2022DigGAN: Discriminator gradIent Gap Regularization for GAN Training with Limited Datacited by 7
  7. 2022DEQGAN: Learning the Loss Function for PINNs with Generative Adversarial Networkscited by 4
  8. 2021Guiding GANs: How to control non-conditional pre-trained GANs for conditional image generationcited by 1
  9. 2022Learning Distributions by Generative Adversarial Networks: Approximation and Generalizationcited by 1
  10. 2025Performative Drift Resistant Classification Using Generative Domain Adversarial Networkscited by 1
  11. 2025MAGIC: Few-Shot Mask-Guided Anomaly Inpainting with Prompt Perturbation, Spatially Adaptive Guidance, and Context Awarenesscited by 1
  12. 2022Effective Shortcut Technique for GANcited by 0
6 more
  1. 2022How to train your draGAN: A task oriented solution to imbalanced classificationcited by 0
  2. 2023Multispectral Contrastive Learning with Viewmaker Networkscited by 0
  3. 2023Flooding Regularization for Stable Training of Generative Adversarial Networkscited by 0
  4. 2024On depth prediction for autonomous driving using self-supervised learningcited by 0
  5. 2024Multi-Armed Bandit Approach for Optimizing Training on Synthetic Datacited by 0
  6. 2026A Wasserstein GAN-based climate scenario generator for risk management and insurance: the case of soil subsidencecited by 0