mlsanityData to 14 Oct 2026

DigGAN: Discriminator gradIent Gap Regularization for GAN Training with Limited Data

arXiv 2211.14694 ↗First posted 27 Nov 2022cs.LG cs.CVcited by 7

Not on the map. Why?

Abstract (arXiv)

Generative adversarial nets (GANs) have been remarkably successful at learning to sample from distributions specified by a given dataset, particularly if the given dataset is reasonably large compared to its dimensionality. However, given limited data, classical GANs have struggled, and strategies like output-regularization, data-augmentation, use of pre-trained models and pruning have been shown to lead to improvements. Notably, the applicability of these strategies is 1) often constrained to particular settings, e.g., availability of a pretrained GAN; or 2) increases training time, e.g., when using pruning. In contrast, we propose a Discriminator gradIent Gap regularized GAN (DigGAN) formulation which can be added to any existing GAN. DigGAN augments existing GANs by encouraging to narrow the gap between the norm of the gradient of a discriminator's prediction w.r.t. real images and w.r.t. the generated samples. We observe this formulation to avoid bad attractors within the GAN loss landscape, and we find DigGAN to significantly improve the results of GAN training when limited data is available. Code is available at urlhttps://github.com/AilsaF/DigGAN.

Key facts. Cited by 7 papers in the map (data through 14 Oct 2026). Builds on 55 of its 77 references that are in the map. First posted on arXiv 27 Nov 2022.

Builds on

55 of 77 references.

References · 55

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 7,005
  3. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 6,013
  4. 2016Layer Normalizationcited by 5,669
  5. 2016Image-to-Image Translation with Conditional Adversarial Networkscited by 4,527
  6. 2017Progressive Growing of GANs for Improved Quality, Stability, and Variationcited by 2,988
  7. 2016Improved Techniques for Training GANscited by 2,970
  8. 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,626
  9. 2017Improved Training of Wasserstein GANscited by 2,235
  10. 2015Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networkscited by 2,054
  11. 2018Spectral Normalization for Generative Adversarial Networkscited by 1,636
  12. 2016Least Squares Generative Adversarial Networkscited by 1,001
38 more
  1. 2018Self-Attention Generative Adversarial Networkscited by 809
  2. 2018Which Training Methods for GANs do actually Converge?cited by 679
  3. 2017Wasserstein GANcited by 639
  4. 2016f-GAN: Training Generative Neural Samplers using Variational Divergence Minimizationcited by 506
  5. 2020Differentiable Augmentation for Data-Efficient GAN Trainingcited by 363
  6. 2016Adversarial Feature Learningcited by 350
  7. 2016NIPS 2016 Tutorial: Generative Adversarial Networkscited by 285
  8. 2017MMD GAN: Towards Deeper Understanding of Moment Matching Networkcited by 254
  9. 2018The relativistic discriminator: a key element missing from standard GANcited by 238
  10. 2021Few-shot Image Generation via Cross-domain Correspondencecited by 207
  11. 2019U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translationcited by 186
  12. 2017Generate To Adapt: Aligning Domains using Generative Adversarial Networkscited by 183
  13. 2019Consistency Regularization for Generative Adversarial Networkscited by 136
  14. 2019Self-Ensembling with GAN-based Data Augmentation for Domain Adaptation in Semantic Segmentationcited by 126
  15. 2017BEGAN: Boundary Equilibrium Generative Adversarial Networkscited by 121
  16. 2019Max-Sliced Wasserstein Distance and its use for GANscited by 119
  17. 2017Stabilizing Training of Generative Adversarial Networks through Regularizationcited by 118
  18. 2016Neural Photo Editing with Introspective Adversarial Networkscited by 109
  19. 2018Generative Modeling using the Sliced Wasserstein Distancecited by 108
  20. 2021Regularizing Generative Adversarial Networks under Limited Datacited by 107
  21. 2017PacGAN: The power of two samples in generative adversarial networkscited by 104
  22. 2021Ensembling Off-the-shelf Models for GAN Trainingcited by 86
  23. 2018The Unusual Effectiveness of Averaging in GAN Trainingcited by 73
  24. 2020Improved Consistency Regularization for GANscited by 69
  25. 2019Sliced Wasserstein Generative Modelscited by 61
  26. 2021Data-Efficient Instance Generation from Instance Discriminationcited by 58
  27. 2021Enjoy Your Editing: Controllable GANs for Image Editing via Latent Space Navigationcited by 53
  28. 2021Deceive D: Adaptive Pseudo Augmentation for GAN Training with Limited Datacited by 53
  29. 2018On the Convergence and Robustness of Training GANs with Regularized Optimal Transportcited by 51
  30. 2019Detecting Overfitting of Deep Generative Networks via Latent Recoverycited by 44
  31. 2017Approximation and Convergence Properties of Generative Adversarial Learningcited by 28
  32. 2021Data-Efficient GAN Training Beyond (Just) Augmentations: A Lottery Ticket Perspectivecited by 25
  33. 2022Collapse by Conditioning: Training Class-conditional GANs with Limited Datacited by 25
  34. 2017Fisher GANcited by 22
  35. 2018BourGAN: Generative Networks with Metric Embeddingscited by 22
  36. 2017McGan: Mean and Covariance Feature Matching GANcited by 21
  37. 2017On Convergence and Stability of GANscited by 18
  38. 2020Towards a Better Global Loss Landscape of GANscited by 17

50 most cited of 55 All 55 (JSON)

Built on by

Cited by year

Citing papers per year: 2023: 2; 2024: 2; 2025: 3.

7 papers.

  1. 2023A Survey on Generative Modeling with Limited Data, Few Shots, and Zero Shotcited by 19
  2. 2023Image Synthesis under Limited Data: A Survey and Taxonomycited by 9
  3. 2024Generative Artificial Intelligence Meets Synthetic Aperture Radar: A Surveycited by 9
  4. 2025Phi-GAN: Physics-Inspired GAN for Generating SAR Images Under Limited Datacited by 4
  5. 2024CHAIN: Enhancing Generalization in Data-Efficient GANs via lipsCHitz continuity constrAIned Normalizationcited by 2
  6. 2025Style Quantization for Data-Efficient GAN Trainingcited by 0
  7. 2025Wind Power Scenario Generation based on the Generalized Dynamic Factor Model and Generative Adversarial Networkcited by 0