Hadamard Wirtinger Flow for Sparse Phase Retrieval

TL;DR

This paper introduces Hadamard Wirtinger flow (HWF), a gradient descent method for sparse phase retrieval that efficiently recovers support and signals with fewer measurements, adapting to sparsity without explicit regularization.

Contribution

The paper proposes HWF, a novel gradient-based algorithm that recovers sparse signals from magnitude-only measurements with improved sample complexity and no need for regularization.

Findings

Single-step HWF recovers support with near-linear sample complexity.

HWF adapts to signal sparsity without explicit regularization.

Numerical results show HWF outperforms existing methods in measurement efficiency.

Abstract

We consider the problem of reconstructing an $n$-dimensional $k$-sparse signal from a set of noiseless magnitude-only measurements. Formulating the problem as an unregularized empirical risk minimization task, we study the sample complexity performance of gradient descent with Hadamard parametrization, which we call Hadamard Wirtinger flow (HWF). Provided knowledge of the signal sparsity $k$, we prove that a single step of HWF is able to recover the support from $k(x^*_{max})^{-2}$ (modulo logarithmic term) samples, where $x^*_{max}$ is the largest component of the signal in magnitude. This support recovery procedure can be used to initialize existing reconstruction methods and yields algorithms with total runtime proportional to the cost of reading the data and improved sample complexity, which is linear in $k$ when the signal contains at least one large component. We numerically investigate the performance of HWF at convergence and show that, while not requiring any explicit form of regularization nor knowledge of $k$, HWF adapts to the signal sparsity and reconstructs sparse signals with fewer measurements than existing gradient based methods.