We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Mathematics > Numerical Analysis

Title: The regularization continuation method with an adaptive time step control for linearly constrained optimization problems

Abstract: This paper considers the regularization continuation method and the trust-region updating strategy for the optimization problem with linear equality constraints.The proposed method utilizes the linear conservation law of the regularization continuation method such that it does not need to compute the correction step for preserving the feasibility other than the previous continuation methods and the quasi-Newton updating formulas for the linearly constrained optimization problem. Moreover, the new method uses the special limited-memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS) formula as the preconditioning technique to improve its computational efficiency in the well-posed phase, and it uses the inverse of the regularized two-sided projection of the Lagrangian Hessian as the pre-conditioner to improve its robustness. Numerical results also show that the new method is more robust and faster than the traditional optimization method such as the alternating direction method of multipliers (ADMM),the sequential quadratic programming (SQP) method (the built-in subroutine fmincon.m of the MATLAB2020a environment), and the recent continuation method (Ptctr). The computational time of the new method is about 1/3 of that of SQP (fmincon.m). Finally, the global convergence analysis of the new method is also given.
Comments: arXiv admin note: substantial text overlap with arXiv:2101.07055, arXiv:2012.14808
Subjects: Numerical Analysis (math.NA); Optimization and Control (math.OC)
Cite as: arXiv:2106.01122 [math.NA]
  (or arXiv:2106.01122v3 [math.NA] for this version)

Submission history

From: Xin-Long Luo [view email]
[v1] Wed, 2 Jun 2021 12:46:53 GMT (62kb,D)
[v2] Tue, 6 Jul 2021 03:40:22 GMT (71kb,D)
[v3] Fri, 8 Apr 2022 08:20:10 GMT (72kb,D)

Link back to: arXiv, form interface, contact.