Revision as of 18:03, 31 December 2021 edit Saint Aardvark (talk \| contribs) 334 edits →Generative Adversarial Networks (GAN) Tag: 2017 wikitext editor ← Previous edit		Revision as of 00:53, 12 April 2022 edit undo Michaelwallace22 (talk \| contribs) Extended confirmed users 131,777 edits mNo edit summary Tag: 2017 wikitext editor Next edit →
Line 108: === Generative Adversarial Networks (GAN) === VQ has been used to quantize a feature representation layer in the discriminator of [[Generative ~~Adversarial~~adversarial ~~Networks~~network]]s. The feature quantization (FQ) technique performs implicit feature matching.<ref>Feature Quantization Improves GAN Training https://arxiv.org/abs/2004.02088</ref> It improves the GAN training, and yields an improved performance on a variety of popular GAN models: BigGAN for image generation, StyleGAN for face synthesis, and U-GAT-IT for unsupervised image-to-image translation. == See also ==

Vector quantization: Difference between revisions