Convolutional Phase Retrieval via Gradient Descent

DOI10.1109/TIT.2019.2950717MaRDI QIDQ5123818zbMATH OpenOpenAlexFDO

Authors

Publication date 29 September 2020

Published in IEEE Transactions on Information Theory (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1712.00716

Mathematics Subject Classification ID

Signal theory (characterization, reconstruction, filtering, etc.) (94A12)

Abstract: We study the convolutional phase retrieval problem, of recovering an unknown signal $mathbfxinmathbb{C}^n$ from $m$ measurements consisting of the magnitude of its cyclic convolution with a given kernel $mathbfainmathbb{C}^m$. This model is motivated by applications such as channel estimation, optics, and underwater acoustic communication, where the signal of interest is acted on by a given channel/filter, and phase information is difficult or impossible to acquire. We show that when $mathbfa$ is random and the number of observations $m$ is sufficiently large, with high probability $mathbfx$ can be efficiently recovered up to a global phase shift using a combination of spectral initialization and generalized gradient descent. The main challenge is coping with dependencies in the measurement operator. We overcome this challenge by using ideas from decoupling theory, suprema of chaos processes and the restricted isometry property of random circulant matrices, and recent analysis of alternating minimization methods.

Recommendations

• A flexible convex relaxation for phase retrieval
• Gradient descent with random initialization: fast global convergence for nonconvex phase retrieval
• PhaseEqual: Convex Phase Retrieval via Alternating Direction Method of Multipliers
• Phase Retrieval Using Alternating Minimization
• Provable Phase Retrieval with Mirror Descent
• PhaseMax: Convex Phase Retrieval via Basis Pursuit
• Simultaneous phase retrieval and blind deconvolution via convex programming
• Phase Retrieval by Alternating Minimization With Random Initialization
• Scalable incremental nonconvex optimization approach for phase retrieval
• Phase retrieval using alternating minimization in a batch setting

Cited in

(6)

• The global optimization geometry of shallow linear neural networks
• Sparse power factorization: balancing peakiness and sample complexity
• Gradient descent with random initialization: fast global convergence for nonconvex phase retrieval
• Exact Recovery of Multichannel Sparse Blind Deconvolution via Gradient Descent
• Convex and Nonconvex Optimization Are Both Minimax-Optimal for Noisy Blind Deconvolution Under Random Designs
• Stability in phase retrieval: characterizing condition numbers and the optimal vector set

This page was built for publication: Convolutional Phase Retrieval via Gradient Descent

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5123818)