摘要
In this paper,a global optimum-based search strategy is proposed to alleviate the situation that the differential evolution(DE)usually sticks into a stagnation,especially on complex problems.It aims to reconstruct the balance between exploration and exploitation,and improve the search efficiency and solution quality of DE.The proposed method is activated by recording the number of recently consecutive unsuccessful global optimum updates.It takes the feedback from the global optimum,which makes the search strategy not only refine the current solution quality,but also have a change to find other promising space with better individuals.This search strategy is incorporated with various DE mutation strategies and DE variations.The experimental results indicate that the proposed method has remarkable performance in enhancing search efficiency and improving solution quality.
基金
This work was supported by the JSPS KAKENHI(JP17K12751 and JP15K00332).
作者简介
Yang Yu,received the B.E.degree in civil engineering from the Yancheng Institute of Technology,China,in 2014,and M.S.degree from University of Toyama,Japan in 2017.He is currently pursuing the Ph.D.degree at University of Toyama,Toyama,Japan.His research interests include data mining,optimization problems,and nature-inspired algorithms.,e-mail:ntrqz@hotmail.com;Corresponding author:Shangce Gao,(M’13-SM’16)received the Ph.D.degree in innovative life science from University of Toyama,Japan in 2011.He is currently an Associate Professor with the Faculty of Engineering,University of Toyama,Japan.His current research interests include natureinspired technologies,mobile computing,machine learning,and neural networks for real-world applications.Dr.Gao was a recipient of a Best Paper Award at IEEE International Conference on Progress in Informatics and Computing,the Shanghai Rising-Star Scientist award,the Chen-Guang Scholar of Shanghai award,the Outstanding Academic Performance Award of IEICE,and the Outstanding Academic Achievement Award of IPSJ,e-mail:gaosc@eng.u-toyama.ac.jp;Yirui Wang,received the B.S.and M.S.degrees with the College of Information Sciences and Technology,from Donghua University,China in 2014 and 2017,respectively.He is currently working toward the Ph.D.degree at University of Toyama,Japan.His research interests include computational intelligence,swarm intelligent algorithms,and combinatorial optimizations,e-mail:wyr607@foxmail.com;Yuki Todo,(M’13)received the B.S.degree from Zhejiang University,China,the M.S.degree from Beijing University of Posts and Telecommunications,China,and the D.E.degree from Kanazawa University,Japan,in 1983,1986,and 2005,respectively.From 1987 to 1989,she was an Assistant Professor with the Institute of Microelectronics,Shanghai Jiaotong University,China.From 1989 to 1990,she was a research student at Nagoya University,Nagoya,Japan.From 1990 to 2000,she was a Senior Engineer with Sanwa Newtech Inc.,Miyazaki,Japan.From 2000 to 2011,she worked with Tateyama Systems Institute,Japan.In 2012,she joined Kanazawa University,where she is now an Associate Professor with the School of Electrical and Computer Engineering.Her current research interests include multiple-valued logic,neural networks,and optimization,e-mail:yktodo@ec.t.kanazawa-u.ac.jp
