Group Forward–Backward Orthogonal Matching Pursuit for General Convex Smooth Functions

This paper introduces the Group Forward–Backward Orthogonal Matching Pursuit (Group-FoBa-OMP) algorithm, a novel approach for sparse feature selection. The core innovations of this algorithm include (1) an integrated backward elimination process to correct earlier misidentified groups; (2) a versati...

Full description

Saved in:
Bibliographic Details
Main Authors: Zhongxing Peng, Gengzhong Zheng, Wei Huang
Format: Article
Language:English
Published: MDPI AG 2024-11-01
Series:Axioms
Subjects:
Online Access:https://www.mdpi.com/2075-1680/13/11/774
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846154335334957056
author Zhongxing Peng
Gengzhong Zheng
Wei Huang
author_facet Zhongxing Peng
Gengzhong Zheng
Wei Huang
author_sort Zhongxing Peng
collection DOAJ
description This paper introduces the Group Forward–Backward Orthogonal Matching Pursuit (Group-FoBa-OMP) algorithm, a novel approach for sparse feature selection. The core innovations of this algorithm include (1) an integrated backward elimination process to correct earlier misidentified groups; (2) a versatile convex smooth model that generalizes previous research; (3) the strategic use of gradient information to expedite the group selection phase; and (4) a theoretical validation of its performance in terms of support set recovery, variable estimation accuracy, and objective function optimization. These advancements are supported by experimental evidence from both synthetic and real-world data, demonstrating the algorithm’s effectiveness.
format Article
id doaj-art-8829313e99c244e5a881f1e6793ff184
institution Kabale University
issn 2075-1680
language English
publishDate 2024-11-01
publisher MDPI AG
record_format Article
series Axioms
spelling doaj-art-8829313e99c244e5a881f1e6793ff1842024-11-26T17:50:52ZengMDPI AGAxioms2075-16802024-11-01131177410.3390/axioms13110774Group Forward–Backward Orthogonal Matching Pursuit for General Convex Smooth FunctionsZhongxing Peng0Gengzhong Zheng1Wei Huang2School of Computer Information Engineering, Hanshan Normal University, Chaozhou 521041, ChinaSchool of Computer Information Engineering, Hanshan Normal University, Chaozhou 521041, ChinaSchool of Computer Information Engineering, Hanshan Normal University, Chaozhou 521041, ChinaThis paper introduces the Group Forward–Backward Orthogonal Matching Pursuit (Group-FoBa-OMP) algorithm, a novel approach for sparse feature selection. The core innovations of this algorithm include (1) an integrated backward elimination process to correct earlier misidentified groups; (2) a versatile convex smooth model that generalizes previous research; (3) the strategic use of gradient information to expedite the group selection phase; and (4) a theoretical validation of its performance in terms of support set recovery, variable estimation accuracy, and objective function optimization. These advancements are supported by experimental evidence from both synthetic and real-world data, demonstrating the algorithm’s effectiveness.https://www.mdpi.com/2075-1680/13/11/774compressed sensinggroup orthogonal matching pursuitgroup sparsegroup restricted isometry propertyinstance optimality
spellingShingle Zhongxing Peng
Gengzhong Zheng
Wei Huang
Group Forward–Backward Orthogonal Matching Pursuit for General Convex Smooth Functions
Axioms
compressed sensing
group orthogonal matching pursuit
group sparse
group restricted isometry property
instance optimality
title Group Forward–Backward Orthogonal Matching Pursuit for General Convex Smooth Functions
title_full Group Forward–Backward Orthogonal Matching Pursuit for General Convex Smooth Functions
title_fullStr Group Forward–Backward Orthogonal Matching Pursuit for General Convex Smooth Functions
title_full_unstemmed Group Forward–Backward Orthogonal Matching Pursuit for General Convex Smooth Functions
title_short Group Forward–Backward Orthogonal Matching Pursuit for General Convex Smooth Functions
title_sort group forward backward orthogonal matching pursuit for general convex smooth functions
topic compressed sensing
group orthogonal matching pursuit
group sparse
group restricted isometry property
instance optimality
url https://www.mdpi.com/2075-1680/13/11/774
work_keys_str_mv AT zhongxingpeng groupforwardbackwardorthogonalmatchingpursuitforgeneralconvexsmoothfunctions
AT gengzhongzheng groupforwardbackwardorthogonalmatchingpursuitforgeneralconvexsmoothfunctions
AT weihuang groupforwardbackwardorthogonalmatchingpursuitforgeneralconvexsmoothfunctions