题  目：A class of one-parameter filled functions with the same local minima as the objective function for global optimization problems

报告人:  尚有林 教授

单   位：河南科技大学 

时   间：11月2日 9：00

地   点：数学与统计学院二楼会议室

摘  要：As an effective global optimization method, the filled function algorithm obtains the optimal solution of optimization problems by alternately minimizing the objective function and filled function. One development direction of the filled function method is to construct the filled function with good properties, which directly affects the efficiency of the algorithm and has attracted the attention of scholars. This paper presents a class of one-parameter filled function that is easy to adjust. It is proved theoretically that the filled function is continuously differentiable and has the same local minimizers as the objective function, and these minimizers are all better than the current local minimizer of the objective function. Meanwhile, some natures of the constructed filled function are conducted as followed by a new algorithm is presented. The new filled function algorithm optimizes the iterative framework of the conventional filled function method, improving the efficiency and decreasing the computational cost. The numerical experiments of the new algorithm on several optimization issues are reported with satisfactory computational results, verifying the feasibility and efficiency of the algorithm.

个人简介：尚有林，二级教授，博士生导师，河南省政府特殊津贴专家，河南省高层次人才（C类），曾任河南科技大学数学与统计学院院长，现任河南科技大学运筹与大数据研究院院长，河南省大数据智能分析与优化创新实验室主任，中国运筹学会常务理事，中国运筹学会数学规划分会资深理事，河南省运筹学会理事长。2005年上海大学运筹学与控制论博士毕业，2005年10月-2007年10月同济大学应用数学博士后流动站研究。一直从事最优化理论、算法及其应用研究。在非线性规划、全局优化、系统工程等领域取得多项研究成果，国内外期刊发表论文150多篇。先后到美国、巴西、智利等进行学术访问交流。主持国家自然科学基金面上项目4项、河南省自科基金项目、国际合作项目多项。出版专著1部，主编教材4部。