mlsanityData to 7 Oct 2026

Generalization in Generative Adversarial Networks: A Novel Perspective from Privacy Protection

arXiv 1908.07882 ↗First posted 21 Aug 2019cs.LG cs.CR stat.MLcited by 19

Not on the map. Why?

Abstract (arXiv)

In this paper, we aim to understand the generalization properties of generative adversarial networks (GANs) from a new perspective of privacy protection. Theoretically, we prove that a differentially private learning algorithm used for training the GAN does not overfit to a certain degree, i.e., the generalization gap can be bounded. Moreover, some recent works, such as the Bayesian GAN, can be re-interpreted based on our theoretical insight from privacy protection. Quantitatively, to evaluate the information leakage of well-trained GAN models, we perform various membership attacks on these models. The results show that previous Lipschitz regularization techniques are effective in not only reducing the generalization gap but also alleviating the information leakage of the training dataset.

Key facts. Cited by 19 papers in the map (data through 7 Oct 2026). First posted on arXiv 21 Aug 2019.

Builds on

No reference list yet.

Built on by

Cited by year

Citing papers per year: 2021: 9; 2022: 5; 2023: 2; 2024: 3.

19 papers.

Sort:

By area:

  1. 2021Membership Inference Attacks on Machine Learning: A Surveycited by 306
  2. 2022M^4I: Multi-modal Models Membership Inferencecited by 30
  3. 2021Copyright in Generative Deep Learningcited by 29
  4. 2022On the Privacy Properties of GAN-generated Samplescited by 16
  5. 2022What is a Good Metric to Study Generalization of Minimax Learners?cited by 16
  6. 2021Generative Adversarial Networks: A Survey Towards Private and Secure Applicationscited by 15
  7. 2021Generative Convolution Layer for Image Generationcited by 5
  8. 2024Stability and Generalization in Free Adversarial Trainingcited by 5
  9. 2021A Novel Generator with Auxiliary Branch for Improving GAN Performancecited by 4
  10. 2021Causally Constrained Data Synthesis for Private Data Releasecited by 3
  11. 2024Dual-Model Defense: Safeguarding Diffusion Models from Membership Inference Attacks through Disjoint Data Splittingcited by 3
  12. 2021Generalization of GANs and overparameterized models under Lipschitz continuitycited by 2
7 more
  1. 2023Preserving Privacy in GANs Against Membership Inference Attackcited by 2
  2. 2021Generalization Bounds for Stochastic Gradient Langevin Dynamics: A Unified View via Information Leakage Analysiscited by 1
  3. 2023SoK: Memorisation in machine learningcited by 1
  4. 2024Rethinking Image Skip Connections in StyleGAN2cited by 1
  5. 2021Improving the expressiveness of neural vocoding with non-affine Normalizing Flowscited by 0
  6. 2022Effective Shortcut Technique for GANcited by 0
  7. 2022Improving GANs with a Feature Cycling Generatorcited by 0