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 …
R Muthukumar, J Sulam - SIAM Journal on Mathematics of Data Science, 2023 - SIAM
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 …
X Dai, K Chen, S Tong, J Zhang, X Gao, M Li… - arXiv preprint arXiv …, 2023 - arxiv.org
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 …
P Hand, O Leong, V Voroninski - Communications on Pure and …, 2024 - Wiley Online Library
Advances in compressive sensing (CS) provided reconstruction algorithms of sparse signals from linear measurements with optimal sample complexity, but natural extensions of this …