Exponential convergence of a proximal projection neural network for mixed variational inequalities and applications

被引：30

作者：

Ju, Xingxing ^{[1
]}

Che, Hangjun ^{[1
]}

Li, Chuandong ^{[1
]}

He, Xing ^{[1
]}

Feng, Gang ^{[2
]}

机构：

[1] Southwest Univ, Sch Elect & Informat Engn, Chongqing Key Lab Nonlinear Circuits & Intelligen, Chongqing 400715, Peoples R China

[2] City Univ Hong Kong, Dept Biomed Engn, Hong Kong, Peoples R China

来源：

NEUROCOMPUTING | 2021年 / 454卷

基金：

中国国家自然科学基金;

关键词：

Proximal projection neural networks; Mixed variational inequalities; Global exponential stability; Sparse recovery problems; Min-max problems; ALGORITHM; STABILITY; RECOVERY;

D O I：

10.1016/j.neucom.2021.04.059

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This paper proposes a novel proximal projection neural network (PPNN) to deal with mixed variational inequalities. It is shown that the PPNN has a unique continuous solution under the condition of Lipschitz continuity and that the trajectories of the PPNN converge to the unique equilibrium solution exponentially under some mild conditions. In addition, we study the influence of different parameters on the convergence rate. Furthermore, the proposed PPNN is applied in solving nonlinear complementarity problems, min-max problems, sparse recovery problems and classification and feature selection problems. Finally, numerical and experimental examples are presented to validate the effectiveness of the proposed neurodynamic network. (c) 2021 Elsevier B.V. All rights reserved.

引用

页码：54 / 64

页数：11

共 39 条

[1] A qualitative mathematical analysis of a class of linear variational inequalities via semi-complementarity problems: applications in electronics [J].