Generative adversarial networks (GANs) have drawn enormous attention due to their simple
yet effective training mechanism and superior image generation quality. With the ability to …

GAN inversion aims to invert a given image back into the latent space of a pretrained GAN
model so that the image can be faithfully reconstructed from the inverted code by the …

Modern practice for training classification deepnets involves a terminal phase of training
(TPT), which begins at the epoch where training error first vanishes. During TPT, the training …

Image denoising is a well-known and well studied problem, commonly targeting a
minimization of the mean squared error (MSE) between the outcome and the original image …

In recent years, there have been significant advances in the use of deep learning methods in
inverse problems such as denoising, compressive sensing, inpainting, and super-resolution …

We study approximation of probability measures supported on n-dimensional manifolds
embedded in R^ m by injective flows—neural networks composed of invertible flows and …

Injectivity plays an important role in generative models where it enables inference; in inverse
problems and compressed sensing with generative priors it is a precursor to well …

This work studies the adversarial robustness of parametric functions composed of a linear
predictor and a nonlinear representation map. Our analysis relies on sparse local …

Autoencoding has achieved great empirical success as a framework for learning generative
models for natural images. Autoencoders often use generic deep networks as the encoder …

Advances in compressive sensing (CS) provided reconstruction algorithms of sparse signals
from linear measurements with optimal sample complexity, but natural extensions of this …