PtyGenography: using generative models for regularization of the phase retrieval problem

Selin Aslan, Tristan van Leeuwen, Allard Mosk, Palina Salanevich

Session: General

Keywords: phase retrieval, inverse problems, regularization, generative priors

TL;DR: The paper compares the behavior of classical and generative approaches to the phase retrieval problem under different noise levels and proposes a conbined approach that mitigates the overfitting of the solution to the generative model.

Abstract: In phase retrieval and similar inverse problems, the stability of solutions across different noise levels is crucial for applications. One approach to promote it is using signal priors in a form of a generative model as a regularization, at the expense of introducing a bias in the reconstruction. In this paper, we explore and compare the reconstruction properties of classical and generative inverse problem formulations. We propose a new unified reconstruction approach that mitigates overfitting to the generative model for varying noise levels.

Submission Number: 38