Variational Phase Retrieval with globally convergent preconditioned Proximal AlgorithmMS49

We reformate a general phase retrieval (PR) problem to explicitly contain a Lipchitz differentiable term. The model can be efficiently solved via Partially Preconditioned Proximal Alternating Linearized Minimization (P3ALM). Thanks to the Lipchitz term, we prove the global convergence of P3ALM. We conduct experiments on a variety of PR sampling schemes to show the effectiveness of the proposed method. Finally, some empirical observations are drawn to better characterize the PR problems.

This presentation is part of Minisymposium “MS49 - Image Restoration, Enhancement and Related Algorithms (4 parts)
organized by: Weihong Guo (Case Western Reserve University) , Ke Chen (University of Liverpool) , Xue-Cheng Tai (Hong Kong Baptist University) , Guohui Song (Clarkson University) .

Yifei Lou (University of Texas at Dallas)
image reconstruction, inverse problems, nonlinear optimization