文章基本信息

标题：AN EFFICIENT ONE-LAYER RECURRENT NEURAL NETWORK FOR SOLVING A CLASS OF NONSMOOTH PSEUDOCONVEX OPTIMIZATION PROBLEMS
本地全文：下载
作者：M. J. EBADI ; M. M. HOSSEINI ; S. M. KARBASSI 等
期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2018
卷号：96
期号：7
出版社：Journal of Theoretical and Applied
摘要：In this paper, an efficient one-layer recurrent neural network model which is differential inclusion-based is proposed for solving nonsmooth pseudoconvex optimization problems subject to linear equality constraints. The optimal solution of the original optimization problem is proven to be equivalent with the equilibrium point of the proposed neural network. In addition, the stability of the proposed neural network in the Lyapunov sense and globally convergence to an optimal solution are proven. Some illustrative examples are given to show the effectiveness of the proposed neural network. In addition, an application for condition number optimization is discussed.
关键词：Differential inclusion-based method; Nonsmooth optimization; Recurrent neural network; Lyapunov stability.