Scholarlink

<p>In this paper, we exploit some properties of points in a neighborhood of the solution set of degenerate optimization problems. Combining these facts with the boundedness of the inverse of regularized Jacobian matrix arising in a mixed logarithmic barrier-augmented lagrangian method, we propose an updating rule for parameters of a mixed logarithmic barrier-augmented Lagrangian algorithm. The superlinear convergence of this algorithm is then proved without any constraint qualification. Numerical results on degenerate problems are also presented to confirm theoretical results.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Convergence analysis of a mixed logarithmic barrier-augmented Lagrangian algorithm without constraint qualification

  • 期刊论文

  • 发表日期：

    2025年5月13日

查看全文
  • Tran Ngoc Nguyen

摘要

In this paper, we exploit some properties of points in a neighborhood of the solution set of degenerate optimization problems. Combining these facts with the boundedness of the inverse of regularized Jacobian matrix arising in a mixed logarithmic barrier-augmented lagrangian method, we propose an updating rule for parameters of a mixed logarithmic barrier-augmented Lagrangian algorithm. The superlinear convergence of this algorithm is then proved without any constraint qualification. Numerical results on degenerate problems are also presented to confirm theoretical results.