<p>This paper introduces a relaxed iterative method for approximating solutions to bilevel variational inequality problems (BVIP) in real Hilbert spaces framework. The mathematical structure addressed herein involves two variational inequalities: one constrained by the other, where the mapping in the lower level is pseudomonotone uniformly continuous. We propose an iterative algorithm that specifically addresses cases where the lower-level mapping is pseudomonotone and uniformly continuous. We then establish strong convergence of the generated sequence to a unique solution of the BVIP using an Armijo-line search procedure. We also provide numerical illustrations in infinite-dimensional Hilbert space settings, including a comparative analysis with previous methods, demonstrating that the proposed approach achieves faster convergence and requires less computational time. A significant contribution of this work is the removal of restrictive assumptions commonly found in existing literature, such as weak sequential continuity and other specific conditions for the lower-level mapping. Furthermore, the method improves computational efficiency by reducing the number of metric projections required to just one per iteration.</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

A relaxed iterative method for approximating solutions of pseudomonotone hierarchical variational inequality problems in Hilbert spaces

  • Kibrom Gebrehiwot Gebremeskel,
  • Teklebrhan Guesh Tewele,
  • Rabian Wangkeeree,
  • Yirga Abebe Belay,
  • Tesfalem Hadush Meche

摘要

This paper introduces a relaxed iterative method for approximating solutions to bilevel variational inequality problems (BVIP) in real Hilbert spaces framework. The mathematical structure addressed herein involves two variational inequalities: one constrained by the other, where the mapping in the lower level is pseudomonotone uniformly continuous. We propose an iterative algorithm that specifically addresses cases where the lower-level mapping is pseudomonotone and uniformly continuous. We then establish strong convergence of the generated sequence to a unique solution of the BVIP using an Armijo-line search procedure. We also provide numerical illustrations in infinite-dimensional Hilbert space settings, including a comparative analysis with previous methods, demonstrating that the proposed approach achieves faster convergence and requires less computational time. A significant contribution of this work is the removal of restrictive assumptions commonly found in existing literature, such as weak sequential continuity and other specific conditions for the lower-level mapping. Furthermore, the method improves computational efficiency by reducing the number of metric projections required to just one per iteration.