A Note on the Regularity of Images Generated by Convolutional Neural Networks

04/22/2022
by   Andreas Habring, et al.
0

The regularity of images generated by convolutional neural networks, such as the U-net, generative adversarial networks, or the deep image prior, is analyzed. In a resolution-independent, infinite dimensional setting, it is shown that such images, represented as functions, are always continuous and, in some circumstances, even continuously differentiable, contradicting the widely accepted modeling of sharp edges in images via jump discontinuities. While such statements require an infinite dimensional setting, the connection to (discretized) neural networks used in practice is made by considering the limit as the resolution approaches infinity. As practical consequence, the results of this paper suggest to refrain from basic L2 regularization of network weights in case of images being the network output.

READ FULL TEXT
research
12/19/2022

VC dimensions of group convolutional neural networks

We study the generalization capacity of group convolutional neural netwo...
research
07/24/2020

Interpreting Spatially Infinite Generative Models

Traditional deep generative models of images and other spatial modalitie...
research
01/11/2021

Correlated Weights in Infinite Limits of Deep Convolutional Neural Networks

Infinite width limits of deep neural networks often have tractable forms...
research
03/04/2020

Turbulence Enrichment using Generative Adversarial Networks

Generative Adversarial Networks (GANs) have been widely used for generat...
research
03/03/2020

On the rate of convergence of image classifiers based on convolutional neural networks

Image classifiers based on convolutional neural networks are defined, an...
research
04/24/2021

On the stability of deep convolutional neural networks under irregular or random deformations

The problem of robustness under location deformations for deep convoluti...
research
06/09/2017

Enhancing SDO/HMI images using deep learning

The Helioseismic and Magnetic Imager (HMI) provides continuum images and...

Please sign up or login with your details

Forgot password? Click here to reset