Block-proximal methods with spatially adapted acceleration
From MaRDI portal
Abstract: We study and develop (stochastic) primal--dual block-coordinate descent methods for convex problems based on the method due to Chambolle and Pock. Our methods have known convergence rates for the iterates and the ergodic gap: if each block is strongly convex, if no convexity is present, and more generally a mixed rate for strongly convex blocks, if only some blocks are strongly convex. Additional novelties of our methods include blockwise-adapted step lengths and acceleration, as well as the ability to update both the primal and dual variables randomly in blocks under a very light compatibility condition. In other words, these variants of our methods are doubly-stochastic. We test the proposed methods on various image processing problems, where we employ pixelwise-adapted acceleration.
Recommendations
- Accelerated primal-dual proximal block coordinate updating methods for constrained convex optimization
- Acceleration of the PDHGM on partially strongly convex functions
- Randomized primal-dual proximal block coordinate updates
- A remark on accelerated block coordinate descent for computing the proximity operators of a sum of convex functions
- A first-order primal-dual algorithm for convex problems with applications to imaging
Cites work
- A class of randomized primal-dual algorithms for distributed optimization
- A Fast Iterative Shrinkage-Thresholding Algorithm for Linear Inverse Problems
- A first-order primal-dual algorithm for convex problems with applications to imaging
- A general framework for a class of first order primal-dual algorithms for convex optimization in imaging science
- A primal-dual splitting method for convex optimization involving Lipschitzian, proximable and linear composite terms
- A splitting algorithm for dual monotone inclusions involving cocoercive operators
- Accelerated proximal stochastic dual coordinate ascent for regularized loss minimization
- Acceleration of the PDHGM on partially strongly convex functions
- An algorithm for mean curvature motion
- An iterative thresholding algorithm for linear inverse problems with a sparsity constraint
- ARock: an algorithmic framework for asynchronous parallel coordinate updates
- Bilevel parameter learning for higher-order total variation regularisation models
- Block-wise alternating direction method of multipliers for multiple-block convex programming and beyond
- Convergence analysis of primal-dual algorithms for a saddle-point problem: from contraction perspective
- Coordinate descent algorithms
- Coordinate-friendly structures, algorithms and applications
- Distributed coordinate descent method for learning with big data
- Efficiency of coordinate descent methods on huge-scale optimization problems
- Inertial, corrected, primal-dual proximal splitting
- iPiano: inertial proximal algorithm for nonconvex optimization
- On the convergence of primal-dual hybrid gradient algorithm
- On the ergodic convergence rates of a first-order primal-dual algorithm
- Optimal primal-dual methods for a class of saddle point problems
- Optimization in high dimensions via accelerated, parallel, and proximal coordinate descent
- Parallel coordinate descent methods for big data optimization
- Proximal alternating linearized minimization for nonconvex and nonsmooth problems
- Stochastic primal-dual coordinate method for regularized empirical risk minimization
- Testing and non-linear preconditioning of the proximal point method
- The Primal-Dual Hybrid Gradient Method for Semiconvex Splittings
- Total generalized variation in diffusion tensor imaging
Cited in
(13)- Acceleration of the PDHGM on partially strongly convex functions
- Accelerated primal-dual proximal block coordinate updating methods for constrained convex optimization
- An adaptive block Lanczos algorithm
- Testing and non-linear preconditioning of the proximal point method
- Primal-dual block-proximal splitting for a class of non-convex problems
- An alternative extrapolation scheme of PDHGM for saddle point problem with nonlinear function
- Accelerated block-coordinate relaxation for regularized optimization
- scientific article; zbMATH DE number 434532 (Why is no real title available?)
- Acceleration and global convergence of a first-order primal-dual method for nonconvex problems
- A remark on accelerated block coordinate descent for computing the proximity operators of a sum of convex functions
- Inertial, corrected, primal-dual proximal splitting
- Relaxed Gauss-Newton methods with applications to electrical impedance tomography
- Preconditioned proximal point methods and notions of partial subregularity
This page was built for publication: Block-proximal methods with spatially adapted acceleration
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2323015)
