https://scholars.lib.ntu.edu.tw/handle/123456789/406505
標題: | Frank-Wolfe works for non-Lipschitz continuous gradient objectives: Scalable poisson phase retrieval. | 作者: | Odor, Gergely Yen-Huan Li Yurtsever, Alp Hsieh, Ya-Ping Tran-Dinh, Quoc Halabi, Marwa El Cevher, Volkan |
關鍵字: | Frank-Wolfe algorithm; non-Lipschitz continuous gradient; Phase retrieval; PhaseLift; Poisson noise | 公開日期: | 2016 | 卷: | 2016-May | 起(迄)頁: | 6230-6234 | 來源出版物: | 2016 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016 | 摘要: | 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. © 2016 IEEE. |
描述: | Shanghai, China |
URI: | https://doi.org/10.1109/ICASSP.2016.7472875 https://scholars.lib.ntu.edu.tw/handle/123456789/406505 https://www.scopus.com/inward/record.uri?eid=2-s2.0-84973343777&doi=10.1109%2fICASSP.2016.7472875&partnerID=40&md5=135e51df265aed016f8545c69ee9a961 |
ISSN: | 15206149 | DOI: | 10.1109/ICASSP.2016.7472875 |
顯示於: | 資訊工程學系 |
在 IR 系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。