An analysis of sketched IRLS for accelerated sparse residual regression

This paper studies the problem of sparse residual regression, i.e., learning a linear model using a norm that favors solutions in which the residuals are sparsely distributed. This is a common problem in a wide range of computer vision applications where a linear system has a lot more equations than...

Full description

Saved in:
Bibliographic Details
Main Authors: IWATA, Daichi, WAECHTER, Michael, LIN, Wen-yan, MATSUSHITA, Yasuyuki
Format: text
Language:English
Published: Institutional Knowledge at Singapore Management University 2020
Subjects:
Online Access:https://ink.library.smu.edu.sg/sis_research/6110
https://ink.library.smu.edu.sg/context/sis_research/article/7113/viewcontent/123570596.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Singapore Management University
Language: English
id sg-smu-ink.sis_research-7113
record_format dspace
spelling sg-smu-ink.sis_research-71132021-09-29T12:30:55Z An analysis of sketched IRLS for accelerated sparse residual regression IWATA, Daichi WAECHTER, Michael LIN, Wen-yan MATSUSHITA, Yasuyuki This paper studies the problem of sparse residual regression, i.e., learning a linear model using a norm that favors solutions in which the residuals are sparsely distributed. This is a common problem in a wide range of computer vision applications where a linear system has a lot more equations than unknowns and we wish to find the maximum feasible set of equations by discarding unreliable ones. We show that one of the most popular solution methods, iteratively reweighted least squares (IRLS), can be significantly accelerated by the use of matrix sketching. We analyze the convergence behavior of the proposed method and show its efficiency on a range of computer vision applications. The source code for this project can be found at https://github.com/Diwata0909/Sketched IRLS. 2020-08-01T07:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/6110 info:doi/doi.org/10.1007/978-3-030-58610-2_36 https://ink.library.smu.edu.sg/context/sis_research/article/7113/viewcontent/123570596.pdf http://creativecommons.org/licenses/by-nc-nd/4.0/ Research Collection School Of Computing and Information Systems eng Institutional Knowledge at Singapore Management University Sparse residual regression L1 minimization Randomized algorithm Matrix sketching Artificial Intelligence and Robotics
institution Singapore Management University
building SMU Libraries
continent Asia
country Singapore
Singapore
content_provider SMU Libraries
collection InK@SMU
language English
topic Sparse residual regression
L1 minimization
Randomized algorithm
Matrix sketching
Artificial Intelligence and Robotics
spellingShingle Sparse residual regression
L1 minimization
Randomized algorithm
Matrix sketching
Artificial Intelligence and Robotics
IWATA, Daichi
WAECHTER, Michael
LIN, Wen-yan
MATSUSHITA, Yasuyuki
An analysis of sketched IRLS for accelerated sparse residual regression
description This paper studies the problem of sparse residual regression, i.e., learning a linear model using a norm that favors solutions in which the residuals are sparsely distributed. This is a common problem in a wide range of computer vision applications where a linear system has a lot more equations than unknowns and we wish to find the maximum feasible set of equations by discarding unreliable ones. We show that one of the most popular solution methods, iteratively reweighted least squares (IRLS), can be significantly accelerated by the use of matrix sketching. We analyze the convergence behavior of the proposed method and show its efficiency on a range of computer vision applications. The source code for this project can be found at https://github.com/Diwata0909/Sketched IRLS.
format text
author IWATA, Daichi
WAECHTER, Michael
LIN, Wen-yan
MATSUSHITA, Yasuyuki
author_facet IWATA, Daichi
WAECHTER, Michael
LIN, Wen-yan
MATSUSHITA, Yasuyuki
author_sort IWATA, Daichi
title An analysis of sketched IRLS for accelerated sparse residual regression
title_short An analysis of sketched IRLS for accelerated sparse residual regression
title_full An analysis of sketched IRLS for accelerated sparse residual regression
title_fullStr An analysis of sketched IRLS for accelerated sparse residual regression
title_full_unstemmed An analysis of sketched IRLS for accelerated sparse residual regression
title_sort analysis of sketched irls for accelerated sparse residual regression
publisher Institutional Knowledge at Singapore Management University
publishDate 2020
url https://ink.library.smu.edu.sg/sis_research/6110
https://ink.library.smu.edu.sg/context/sis_research/article/7113/viewcontent/123570596.pdf
_version_ 1770575831601512448