Frank-Wolfe works for non-Lipschitz continuous gradient objectives: Scalable poisson phase retrieval
Gergely Ódor, Yen-Huan Li, Alp Yurtsever, Ya-Ping Hsieh, Quoc Tran-Dinh, Marwa El Halabi, Volkan Cevher
Abstract
We study a phase retrieval problem in the Poisson noise model. Motivated by the PhaseLift approach, we approximate the maximum-likelihood estimator by solving a convex program with a nuclear norm constraint. While the Frank-Wolfe algorithm, together with the Lanczos method, can efficiently deal with nuclear norm constraints, our objective function does not have a Lipschitz continuous gradient, and hence existing convergence guarantees for the Frank-Wolfe algorithm do not apply. In this paper, we show that the Frank-Wolfe algorithm works for the Poisson phase retrieval problem, and has a global convergence rate of O(1/t), where t is the iteration counter. We provide rigorous theoretical guarantee and illustrating numerical results.
BibTeX
@inproceedings{icassp2016_frankwolfeworksf,
title = {Frank-Wolfe works for non-Lipschitz continuous gradient objectives: Scalable poisson phase retrieval},
author = {Gergely Ódor and Yen-Huan Li and Alp Yurtsever and Ya-Ping Hsieh and Quoc Tran-Dinh and Marwa El Halabi and Volkan Cevher},
booktitle = {ICASSP 2016},
year = {2016}
}