Abstract:The phase retrieval problem, where one aims to recover a complex-valued image from far-field intensity measurements, is a classic problem encountered in a range of imaging applications. Modern phase retrieval approaches usually rely on gradient descent methods in a nonlinear minimization framework. Calculating closed-form gradients for use in these methods is tedious work, and formulating second order derivatives is even more laborious. Additionally, second order techniques often require the storage and inversion of large matrices of partial derivatives, with memory requirements that can be prohibitive for data-rich imaging modalities. We use a reverse-mode automatic differentiation (AD) framework to implement an efficient matrix-free version of the Levenberg-Marquardt (LM) algorithm, a longstanding method that finds popular use in nonlinear least-square minimization problems but which has seen little use in phase retrieval. Furthermore, we extend the basic LM algorithm so that it can be applied for general constrained optimization problems beyond just the least-square applications. Since we use AD, we only need to specify the physics-based forward model for a specific imaging application; the derivative terms are calculated automatically through matrix-vector products, without explicitly forming any large Jacobian or Gauss-Newton matrices. We demonstrate that this algorithm can be used to solve both the unconstrained ptychographic object retrieval problem and the constrained "blind" ptychographic object and probe retrieval problems, under both the Gaussian and Poisson noise models, and that this method outperforms best-in-class first-order ptychographic reconstruction methods: it provides excellent convergence guarantees with (in many cases) a superlinear rate of convergence, all with a computational cost comparable to, or lower than, the tested first-order algorithms.

Wirtinger gradient descent methods for low-dose Poisson phase retrieval

Riemannian Optimization for Phase Retrieval from Masked Fourier Measurements

Wirtinger Gradient Descent Optimization for Reducing Gaussian Noise in Lensless Microscopy

PhaseMax: Stable Guarantees from Noisy Sub-Gaussian Measurements

PGPAL: A Monotonic Iterative Algorithm for Phase-Retrieval Under the Presence of Poisson-Gaussian Noise

Robust Gradient Descent for Phase Retrieval

Low-light phase retrieval with implicit generative priors

Total Variation-Based Phase Retrieval for Poisson Noise Removal.

Optimizing Double-Phase Method Based on Gradient Descent Algorithm with Complex Spectrum Loss Function

Phase Diverse Phase Retrieval for Microscopy: Comparison of Gaussian and Poisson Approaches

Influence of loss function and electron dose on ptychography of 2D materials using the Wirtinger flow

Accelerated Wirtinger Flow With Score-Based Image Priors for Holographic Phase Retrieval in Poisson-Gaussian Noise Conditions

Maximum-likelihood estimation in ptychography in the presence of Poisson-Gaussian noise statistics

Denoising Poisson phaseless measurements via orthogonal dictionary learning

Fourier Ptychographic Reconstruction Using Poisson Maximum Likelihood and Truncated Wirtinger Gradient

A matrix-free Levenberg-Marquardt algorithm for efficient ptychographic phase retrieval

Poisson-Gaussian Holographic Phase Retrieval with Score-based Image Prior

Conjugate Gradient Method for Phase Retrieval Based on the Wirtinger Derivative

A Continuous-Time Mirror Descent Approach to Sparse Phase Retrieval

An Undersampled Phase Retrieval Algorithm Via Gradient Iteration