Accelerated variational PDEs for efficient solution of regularized inversion problems
From MaRDI portal
accelerated gradient descentBeltrami regularizationimage deblurringimage denoisingimage restorationNesterov accelerationnonlinear wave equationsPDE accelerationtotal variation
Smoothness and regularity of solutions to PDEs (35B65) PDEs in connection with information and communication (35Q94) Acceleration of convergence in numerical analysis (65B99) Numerical optimization and variational techniques (65K10) Finite difference methods for initial value and initial-boundary value problems involving PDEs (65M06) Numerical methods for inverse problems for initial value and initial-boundary value problems involving PDEs (65M32) Image processing (compression, reconstruction, etc.) in information and communication theory (94A08)
Abstract: We further develop a new framework, called PDE Acceleration, by applying it to calculus of variations problems defined for general functions on , obtaining efficient numerical algorithms to solve the resulting class of optimization problems based on simple discretizations of their corresponding accelerated PDE's. While the resulting family of PDE's and numerical schemes are quite general, we give special attention to their application for regularized inversion problems, with particular illustrative examples on some popular image processing applications. The method is a generalization of momentum, or accelerated, gradient descent to the PDE setting. For elliptic problems, the descent equations are a nonlinear damped wave equation, instead of a diffusion equation, and the acceleration is realized as an improvement in the CFL condition from (for diffusion) to (for wave equations). We work out several explicit as well as a semi-implicit numerical schemes, together with their necessary stability constraints, and include recursive update formulations which allow minimal-effort adaptation of existing gradient descent PDE codes into the accelerated PDE framework. We explore these schemes more carefully for a broad class of regularized inversion applications, with special attention to quadratic, Beltrami, and Total Variation regularization, where the accelerated PDE takes the form of a nonlinear wave equation. Experimental examples demonstrate the application of these schemes for image denoising, deblurring, and inpainting, including comparisons against Primal Dual, Split Bregman, and ADMM algorithms.
Recommendations
- On the acceleration of optimal regularization algorithms for linear ill-posed inverse problems
- Conjugate gradient based acceleration for inverse problems
- A new acceleration of variational iteration method for initial value problems
- Accelerated projected steepest descent method for nonlinear inverse problems with sparsity constraints
- Accelerated schemes for a class of variational inequalities
- Solution paths of variational regularization methods for inverse problems
- Fast fully iterative Newton-type methods for inverse problems
- An accelerated homotopy-perturbation-Kaczmarz method for solving nonlinear inverse problems
- Regularized fixed-point iterations for nonlinear inverse problems
- Acceleration techniques for regularized Newton methods applied to electromagnetic inverse medium scattering problems
Cites work
- A first-order primal-dual algorithm for convex problems with applications to imaging
- A general framework for low level vision
- A variational perspective on accelerated methods in optimization
- Accelerated Optimization in the PDE Framework Formulations for the Active Contour Case
- Accelerated Optimization in the PDE Framework: Formulations for the Manifold of Diffeomorphisms
- Active contours without edges
- An introduction to continuous optimization for imaging
- Backtracking strategies for accelerated descent methods with smooth composite objectives
- Cyclic schemes for PDE-based image analysis
- Damped second order flow applied to image denoising
- scientific article; zbMATH DE number 3850830 (Why is no real title available?)
- scientific article; zbMATH DE number 2107836 (Why is no real title available?)
- Large-scale machine learning with stochastic gradient descent
- Nonlinear total variation based noise removal algorithms
- Some methods of speeding up the convergence of iteration methods
- The gradient and heavy ball with friction dynamical systems: The quasiconvex case
- THE HEAVY BALL WITH FRICTION METHOD, I. THE CONTINUOUS DYNAMICAL SYSTEM: GLOBAL EXPLORATION OF THE LOCAL MINIMA OF A REAL-VALUED FUNCTION BY ASYMPTOTIC ANALYSIS OF A DISSIPATIVE DYNAMICAL SYSTEM
- The Split Bregman Method for L1-Regularized Problems
Cited in
(10)- Preconditioned accelerated gradient descent methods for locally Lipschitz smooth objectives with applications to the solution of nonlinear PDEs
- PDE acceleration: a convergence rate analysis and applications to obstacle problems
- A generalized computationally efficient inverse characterization approach combining direct inversion solution initialization with gradient-based optimization
- Second-order flows for computing the ground states of rotating Bose-Einstein condensates
- Accelerated Optimization in the PDE Framework: Formulations for the Manifold of Diffeomorphisms
- Accelerated Optimization in the PDE Framework Formulations for the Active Contour Case
- A class of second-order geometric quasilinear hyperbolic PDEs and their application in imaging
- Fast convergence strategy for ambiguous inverse problems based on hierarchical regularization
- A primal-dual level set method for computing geodesic distances
- An accelerated Gauss-Seidel method for inverse modeling
This page was built for publication: Accelerated variational PDEs for efficient solution of regularized inversion problems
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2303258)
