BriefGPT.xyz
May, 2018
散射变换下基于对抗生成网络的反问题
Generative networks as inverse problems with Scattering transforms
HTML
PDF
Tomás Angles, Stéphane Mallat
TL;DR
本研究使用Lipschitz连续的embedding以及wavelet Scattering transform计算深度卷积生成网络，不需要被鉴别器或编码器优化，并且具有类似GAN或VAE的属性。
Abstract
generative adversarial nets
(GANs) and
variational auto-encoders
(VAEs) provide impressive image generations from Gaussian white noise, but the underlying mathematics are not well understood. We compute
→