Abstract 

We introduce a novel deep learning inspired formulation of the phase retrieval problem, which asks to recover a signal from m quadratic observations, under structural assumptions on the underlying signal. As is common in many imaging problems, previous methodologies have considered natural signals as being sparse with respect to a known basis, resulting in the decision to enforce a generic sparsity prior. However, these methods for phase retrieval have encountered possibly fundamental limitations, as no computationally efficient algorithm for sparse phase retrieval has been proven to succeed with fewer than generic measurements, which is larger than the theoretical optimum of In this paper, we propose a new framework for phase retrieval by modeling natural signals as being in the range of a deep generative neural network G :  We introduce an empirical risk formulation that has favorable global geometry for gradient methods, as soon as m = under the model of a d-layer fully-connected neural network with random weights. Specifically, when suitable deterministic conditions on the generator and measurement matrix are met, we construct a descent direction for any point outside of a small neighborhood around the true k-dimensional latent code and a negative multiple thereof. This formulation for structured phase retrieval thus benefits from two effects: generative priors can more tightly represent natural signals than sparsity priors, and this empirical risk formulation can exploit those generative priors at an information theoretically optimal sample complexity, unlike for a sparsity prior. We corroborate these results with experiments showing that exploiting generative models in phase retrieval tasks outperforms both sparse and general phase retrieval methods.