Description
SNGAN, or Spectrally Normalised GAN, is a type of generative adversarial network that uses spectral normalization, a type of weight normalization, to stabilise the training of the discriminator.
Papers Using This Method
Re-GAN: Data-Efficient GANs Training via Architectural Reconfiguration2023-01-01Unrealistic Feature Suppression for Generative Adversarial Networks2021-07-23Exploring The Effect of High-frequency Components in GANs Training2021-03-20Data InStance Prior (DISP) in Generative Adversarial Networks2020-12-08Adaptive Weighted Discriminator for Training Generative Adversarial Networks2020-12-05Improving the Speed and Quality of GAN by Adversarial Training2020-08-07PriorGAN: Real Data Prior for Generative Adversarial Nets2020-06-30Synthesizing Unrestricted False Positive Adversarial Objects Using Generative Models2020-05-19Mimicry: Towards the Reproducibility of GAN Research2020-05-05LOGAN: Latent Optimisation for Generative Adversarial Networks2019-12-02Deep Compressed Sensing2019-05-16Spectral Normalization for Generative Adversarial Networks2018-02-16