Manifold learning by mixture models of vaes for inverse problems

GS Alberti, J Hertrich, M Santacesaria… - Journal of Machine …, 2024 - jmlr.org
Representing a manifold of very high-dimensional data with generative models has been
shown to be computationally efficient in practice. However, this requires that the data …

Importance corrected neural JKO sampling

J Hertrich, R Gruhlke - arXiv preprint arXiv:2407.20444, 2024 - arxiv.org
In order to sample from an unnormalized probability density function, we propose to
combine continuous normalizing flows (CNFs) with rejection-resampling steps based on …

Mixed Noise and Posterior Estimation with Conditional DeepGEM

P Hagemann, J Hertrich, M Casfor… - arXiv preprint arXiv …, 2024 - arxiv.org
Motivated by indirect measurements and applications from nanometrology with a mixed
noise model, we develop a novel algorithm for jointly estimating the posterior and the noise …

Invertible ResNets for Inverse Imaging Problems: Competitive Performance with Provable Regularization Properties

C Arndt, J Nickel - arXiv preprint arXiv:2409.13482, 2024 - arxiv.org
Learning-based methods have demonstrated remarkable performance in solving inverse
problems, particularly in image reconstruction tasks. Despite their success, these …

Manifold Learning and Sparsity Priors for Inverse Problems

S Sciutto - 2024 - tesidottorato.depositolegale.it
In this thesis we investigate two distinct regularizing approaches for solving inverse
problems. The first approach involves assuming that the unknown belongs to a manifold …