当前位置:首页  教学科研

海外英才分论坛学术报告【Zeroing Neural Network (ZNN): A Unified Framework for Solving Time-Varying Problems】

时间:2018-04-23浏览:1249设置

时间:2018年4月24日10:30-11:00

地点:旗山校区数信大楼507

主讲:吉首大学 肖林副教授

主办:数学与信息学院

专家简介:肖林,男,1986年7月生,2014年6月获中山大学博士学位,现任职于吉首大学副教授,在香港理工大学访学。发表学术论文60余篇,SCI期刊论文38篇,IEEE Transactions系列论文8篇(影响因子累计为110.219),CRC专著1部;授权发明专利、实用新型专利和计算机软件著作权各1项;主持国家自然科学基金青年项目、湖南省自然科学基金面上项目和湖南省教育厅优秀青年项目各1项。

报告摘要:Inspired by the negative impact of additive noiseson zeroing neural network (ZNN) for time-varying problems, a unified zeroing neural network (UZNN) is designed and presented to achievenoise suppression and finite-time convergence simultaneously. Compared to the existing ZNN modelonly with finite-time convergence, the proposed UNNN model inherently possessesthe extra robustness property in front of additive noises, in addition to finite-timeconvergence. Furthermore, the design process, theoretical analysis,and numerical verification of the proposed UZNN model are supplied in details. Boththeoretical and numerical results demonstrate the better property of the proposed UZNNmodelfor solving time-varying problems in the presence of additive noises, as comparedwith the ZNN model.

返回原图
/