Nonconvex Demixing From Bilinear Measurements

被引:28
作者
Dong, Jialin [1 ]
Shi, Yuanming [1 ]
机构
[1] ShanghaiTech Univ, Sch Informat Sci & Technol, Shanghai 201210, Peoples R China
关键词
Blind demixing; blind deconvolution; bilinear measurements; nonconvex optimization; Wirtinger flow; regularization-free; statistical and computational guarantee; BLIND DECONVOLUTION;
D O I
10.1109/TSP.2018.2864660
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
We consider the problem of demixing a sequence of source signals from the sum of noisy bilinear measurements. It is a generalized mathematical model for blind demixing with blind deconvolution, which is prevalent across the areas of dictionary learning, image processing, and communications. However, state-of-the-art convex methods for blind demixing via semidefinite programming are computationally infeasible for large-scale problems. Although the existing nonconvex algorithms are able to address the scaling issue, they normally require proper regularization to establish optimality guarantees. The additional regularization yields tedious algorithmic parameters and pessimistic convergence rates with conservative step sizes. To address the limitations of exiting methods, we thus develop a provable nonconvex demixing procedure via Wirtinger flow, much like vanilla gradient descent, to harness the benefits of regularization-free fast convergence rate with aggressive step size and computational optimality guarantees. This is achieved by exploiting the benign geometry of the blind demixing problem, thereby revealing that Wirtinger flow enforces the regularization-free iterates in the region of strong convexity and qualified level of smoothness, where the step size can be chosen aggressively.
引用
收藏
页码:5152 / 5166
页数:15
相关论文
共 23 条
[1]   Blind Deconvolution Using Convex Programming [J].
Ahmed, Ali ;
Recht, Benjamin ;
Romberg, Justin .
IEEE TRANSACTIONS ON INFORMATION THEORY, 2014, 60 (03) :1711-1732
[2]   Living on the edge: phase transitions in convex programs with random data [J].
Amelunxen, Dennis ;
Lotz, Martin ;
McCoy, Michael B. ;
Tropp, Joel A. .
INFORMATION AND INFERENCE-A JOURNAL OF THE IMA, 2014, 3 (03) :224-294
[3]  
[Anonymous], 2015, ADV NEURAL INFORM PR
[4]  
[Anonymous], INF INFERENCE J MAR
[5]  
[Anonymous], 2016, Proc. Mach. Learn. Res. (PMLR)
[6]  
[Anonymous], 2016, Advances in Neural Information Processing Systems
[7]  
[Anonymous], 2015, C LEARNING THEORY
[8]  
[Anonymous], 2018, Applied and Computational Harmonic Analysis
[9]   Fast Convolutional Sparse Coding [J].
Bristow, Hilton ;
Eriksson, Anders ;
Lucey, Simon .
2013 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2013, :391-398
[10]  
Campisi, 2017, Blind image deconvolution: theory and applications