Exact Regularization of Convex Programs
SIAM Journal on Optimization
Sparse reconstruction by separable approximation
IEEE Transactions on Signal Processing
Fixed-Point Continuation for $\ell_1$-Minimization: Methodology and Convergence
SIAM Journal on Optimization
Bregman Iterative Algorithms for $\ell_1$-Minimization with Applications to Compressed Sensing
SIAM Journal on Imaging Sciences
A Singular Value Thresholding Algorithm for Matrix Completion
SIAM Journal on Optimization
Fixed point and Bregman iterative methods for matrix rank minimization
Mathematical Programming: Series A and B
Analysis and Generalizations of the Linearized Bregman Method
SIAM Journal on Imaging Sciences
Nonlinear inverse scale space methods for image restoration
VLSM'05 Proceedings of the Third international conference on Variational, Geometric, and Level Set Methods in Computer Vision
Iterative Regularization and Nonlinear Inverse Scale Space Applied to Wavelet-Based Denoising
IEEE Transactions on Image Processing
Hi-index | 0.00 |
This short article analyzes an interesting property of the Bregman iterative procedure, which is equivalent to the augmented Lagrangian method, for minimizing a convex piece-wise linear function J(x) subject to linear constraints Ax=b. The procedure obtains its solution by solving a sequence of unconstrained subproblems of minimizing $J(x)+\frac{1}{2}\|Ax-b^{k}\|_{2}^{2}$ , where b k is iteratively updated. In practice, the subproblem at each iteration is solved at a relatively low accuracy. Let w k denote the error introduced by early stopping a subproblem solver at iteration k. We show that if all w k are sufficiently small so that Bregman iteration enters the optimal face, then while on the optimal face, Bregman iteration enjoys an interesting error-forgetting property: the distance between the current point $\bar{x}^{k}$ and the optimal solution set X 驴 is bounded by 驴w k+1驴w k 驴, independent of the previous errors w k驴1,w k驴2,驴,w 1. This property partially explains why the Bregman iterative procedure works well for sparse optimization and, in particular, for 驴 1-minimization. The error-forgetting property is unique to J(x) that is a piece-wise linear function (also known as a polyhedral function), and the results of this article appear to be new to the literature of the augmented Lagrangian method.