A preconditioned third-order implicit-explicit algorithm with a difference of varying convex functions and extrapolation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wu, Kelin, Sun, Hongpeng
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909794083799040
author Wu, Kelin
Sun, Hongpeng
author_facet Wu, Kelin
Sun, Hongpeng
contents This paper proposes a novel preconditioned implicit-explicit algorithm enhanced with the extrapolation technique for non-convex optimization problems. The algorithm employs a third-order Adams-Bashforth scheme for the nonlinear and explicit parts and a third-order backward differentiation formula for the implicit part of the gradient flow in variational functions. The proposed algorithm, akin to a generalized difference-of-convex (DC) approach, employs a changing set of convex functions in each iteration. Under the Kurdyka-Łojasiewicz (KL) properties, the global convergence of the algorithm is guaranteed, ensuring that it converges within a finite number of preconditioned iterations. Our numerical experiments, including least squares problems with SCAD regularization and the graphical Ginzburg-Landau model, demonstrate the proposed algorithm's highly efficient performance compared to conventional DC algorithms.
format Preprint
id arxiv_https___arxiv_org_abs_2509_09391
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A preconditioned third-order implicit-explicit algorithm with a difference of varying convex functions and extrapolation
Wu, Kelin
Sun, Hongpeng
Optimization and Control
Numerical Analysis
This paper proposes a novel preconditioned implicit-explicit algorithm enhanced with the extrapolation technique for non-convex optimization problems. The algorithm employs a third-order Adams-Bashforth scheme for the nonlinear and explicit parts and a third-order backward differentiation formula for the implicit part of the gradient flow in variational functions. The proposed algorithm, akin to a generalized difference-of-convex (DC) approach, employs a changing set of convex functions in each iteration. Under the Kurdyka-Łojasiewicz (KL) properties, the global convergence of the algorithm is guaranteed, ensuring that it converges within a finite number of preconditioned iterations. Our numerical experiments, including least squares problems with SCAD regularization and the graphical Ginzburg-Landau model, demonstrate the proposed algorithm's highly efficient performance compared to conventional DC algorithms.
title A preconditioned third-order implicit-explicit algorithm with a difference of varying convex functions and extrapolation
topic Optimization and Control
Numerical Analysis
url https://arxiv.org/abs/2509.09391