Generative networks as inverse problems with Scattering transforms
By Tomás Angles, Stéphane Mallat. 2018
35 citations recorded by the source corpus.
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 deep convolutional network generators by inverting a fixed embedding operator. Therefore, they do not require to be optimized with a discriminator or an encoder. The embedding is Lipschitz continuous to deformations so that generators transform linear interpolations between input white noise vectors into deformations between output images. This embedding is computed with a wavelet Scattering transform. Numerical experiments demonstrate that the resulting Scattering generators have similar properties as GANs or VAEs, without learning a discriminative network or an encoder.
Subfields: Generative Models
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Generative Models
- Eugene Belilovsky
- Simon Lacoste-Julien
- Matthew B. Blaschko
- Gabriel Huang
- Nikos Komodakis
- Edouard Oyallon
- William L. Hamilton
- Gauthier Gidel
Citations in the corpus
- Scattering Networks for Hybrid Representation Learning (2018, methodology): “Concurrent to our work the Scattering Transform was also recently used in a context of generative modeling [2]: it is shown that by inverting the scattering transform, it is pos-”