Improved Convergence Factor of Windowed Anderson Acceleration for Symmetric Fixed-Point Iterations

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Garner, Casey, Lerman, Gilad, Zhang, Teng
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911084930138112
author Garner, Casey
Lerman, Gilad
Zhang, Teng
author_facet Garner, Casey
Lerman, Gilad
Zhang, Teng
contents This paper studies the commonly utilized windowed Anderson acceleration (AA) algorithm for fixed-point methods, $x^{(k+1)}=q(x^{(k)})$. It provides the first proof that when the operator $q$ is linear and symmetric the windowed AA, which uses a sliding window of prior iterates, improves the root-linear convergence factor over the fixed-point iterations. When $q$ is nonlinear, yet has a symmetric Jacobian at a fixed point, a slightly modified AA algorithm is proved to have an analogous root-linear convergence factor improvement over fixed-point iterations. Simulations verify our observations. Furthermore, experiments with different data models demonstrate AA is significantly superior to the standard fixed-point methods for Tyler's M-estimation.
format Preprint
id arxiv_https___arxiv_org_abs_2311_02490
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Improved Convergence Factor of Windowed Anderson Acceleration for Symmetric Fixed-Point Iterations
Garner, Casey
Lerman, Gilad
Zhang, Teng
Numerical Analysis
Optimization and Control
Machine Learning
65F10, 65H10, 68W40
This paper studies the commonly utilized windowed Anderson acceleration (AA) algorithm for fixed-point methods, $x^{(k+1)}=q(x^{(k)})$. It provides the first proof that when the operator $q$ is linear and symmetric the windowed AA, which uses a sliding window of prior iterates, improves the root-linear convergence factor over the fixed-point iterations. When $q$ is nonlinear, yet has a symmetric Jacobian at a fixed point, a slightly modified AA algorithm is proved to have an analogous root-linear convergence factor improvement over fixed-point iterations. Simulations verify our observations. Furthermore, experiments with different data models demonstrate AA is significantly superior to the standard fixed-point methods for Tyler's M-estimation.
title Improved Convergence Factor of Windowed Anderson Acceleration for Symmetric Fixed-Point Iterations
topic Numerical Analysis
Optimization and Control
Machine Learning
65F10, 65H10, 68W40
url https://arxiv.org/abs/2311.02490