Large-scale quasi-Newton trust-region methods with low-dimensional linear equality constraints

被引：6

作者：

Brust, Johannes J. ^{[1
]}

Marcia, Roummel F. ^{[2
]}

Petra, Cosmin G. ^{[3
]}

机构：

[1] Argonne Natl Lab, Lemont, IL 60439 USA

[2] Univ Calif Merced, Merced, CA USA

[3] Lawrence Livermore Natl Lab, Livermore, CA 94550 USA

来源：

COMPUTATIONAL OPTIMIZATION AND APPLICATIONS | 2019年 / 74卷 / 03期

关键词：

Linear equality constraints; Quasi-Newton; L-BFGS; Trust-region algorithm; Compact representation; Eigendecomposition; Shape-changing norm; LIMITED-MEMORY; ALGORITHM; IMPLEMENTATION;

D O I：

10.1007/s10589-019-00127-4

中图分类号：

C93 [管理学]; O22 [运筹学];

学科分类号：

070105 ; 12 ; 1201 ; 1202 ; 120202 ;

摘要：

We propose two limited-memory BFGS (L-BFGS) trust-region methods for large-scale optimization with linear equality constraints. The methods are intended for problems where the number of equality constraints is small. By exploiting the structure of the quasi-Newton compact representation, both proposed methods solve the trust-region subproblems nearly exactly, even for large problems. We derive theoretical global convergence results of the proposed algorithms, and compare their numerical effectiveness and performance on a variety of large-scale problems.

引用

页码：669 / 701

页数：33

共 50 条

[1] Large-scale quasi-Newton trust-region methods with low-dimensional linear equality constraints
Johannes J. Brust
Roummel F. Marcia
Cosmin G. Petra
[J]. Computational Optimization and Applications, 2019, 74 : 669 - 701
[2] A quasi-Newton trust-region method
E. Michael Gertz
[J]. Mathematical Programming, 2004, 100 : 447 - 470
[3] A quasi-Newton trust-region method
Gertz, EM
[J]. MATHEMATICAL PROGRAMMING, 2004, 100 (03) : 447 - 470
[4] AN LDLT TRUST-REGION QUASI-NEWTON METHOD
Brust, Johannes J.
Gill, Philip E.
[J]. SIAM Journal on Scientific Computing, 2024, 46 (05):
[5] Deep Neural Networks Training by Stochastic Quasi-Newton Trust-Region Methods
Yousefi, Mahsa
Martinez, Angeles
[J]. ALGORITHMS, 2023, 16 (10)
[6] Quasi-Newton methods for large-scale electromagnetic inverse problems
Haber, E
[J]. INVERSE PROBLEMS, 2005, 21 (01) : 305 - 323
[7] A PROXIMAL QUASI-NEWTON TRUST-REGION METHOD FOR NONSMOOTH REGULARIZED OPTIMIZATION
Aravkin, Aleksandr Y.
Baraldi, Robert
Orban, Dominique
[J]. SIAM JOURNAL ON OPTIMIZATION, 2022, 32 (02) : 900 - 929
[8] A Class of Diagonal Quasi-Newton Methods for Large-Scale Convex Minimization
Leong, Wah June
[J]. BULLETIN OF THE MALAYSIAN MATHEMATICAL SCIENCES SOCIETY, 2016, 39 (04) : 1659 - 1672
[9] A Class of Diagonal Quasi-Newton Methods for Large-Scale Convex Minimization
Wah June Leong
[J]. Bulletin of the Malaysian Mathematical Sciences Society, 2016, 39 : 1659 - 1672
[10] An Overview of Stochastic Quasi-Newton Methods for Large-Scale Machine Learning
Guo, Tian-De
Liu, Yan
Han, Cong-Ying
[J]. JOURNAL OF THE OPERATIONS RESEARCH SOCIETY OF CHINA, 2023, 11 (02) : 245 - 275

← 1 2 3 4 5 →