应学院邀请，兰州大学郑兵教授将来我院作学术报告。
报告题目：A completely self-scaling G-transformation for weighted least square problems
报告摘要：The G-transformation is an efficient method for solving the weighted least squares problems. However, the underflows and overflows were not considered in the original G-transformation. In order to keep its stability, some specified scaling strategies has been proposed for guarding against the underflows. Note that these specific strategies are not easy to be implemented in actual operations, in this talk we present a completely self-scaling G-transformation (CSSGT) which not only avoids these specified scaling strategies, but maintain the stability of operations. Complexity analysis of our self-scaling G-transformation shows that its cost of computation is less than that of the G-transformation, which implies the high efficiency of our proposed SSGT. The stability of the SSGT was theoretically confirmed by a detailed error analysis. Some numerical experiments are performed to illustrate the effects of the self-scaling strategy.
报告时间：2023年10月29日16:30
报告地点：致勤楼C区101学术报告厅
邀 请 人：孟令胜副教授
届时欢迎广大师生参与交流！
报告人简介
郑兵，兰州大学数学与统计学院教授，博士生导师。长期从事数值代数、神经网络算法的研究工作，负责承担国家自然科学基金面上项目、教育部外国专家重点项目、甘肃省自然科学基金项目等10余项。多次应邀赴美国、日本、西班牙、俄罗斯、印度以及香港、澳门等国家和地区参加学术会议并做学术报告，并先后在印度统计研究所新德里中心和美国Emory大学数学与计算机系做访问学者。迄今已在SIAM J. Matrix Anal. Appl., J. Math. Anal. Appl., J. Optim. Theory Appl., Linear Algebra Appl., J. Multivariate Anal., Adv. Comput. Math., IEEE Trans. Neural Network Learn. Syst., IEEE Trans. Fuzzy Syst., Automatica等国内外重要刊物上发表论文100余篇。2005年荣获甘肃省第十二届高校青年老师成才奖。
甘肃省数学与统计学基础学科研究中心
数学与统计学院
2023年10月26日
