Title :
A small world algorithm for high-dimensional function optimization
Author :
Xiaohu, Li ; Jinhua, Zhang ; Sunan, Wang ; Maolin, Li ; Kunpeng, Li
Author_Institution :
Mech. Eng. Dept. & Eng. Workshop, Xi´´an Jiaotong Univ., Xi´´an, China
Abstract :
In this paper, we describe a new small world optimization algorithm for obtaining satisfactory solution for high-dimensional function. Based on the small world phenomenon which is revealed in Milgram´s sociological experiment, some operators with decimal-coding strategy are proposed, and then an ¿imitated society¿ decimal-coding small world optimization algorithm (DSWOA) is designed to solve high-dimensional function optimization. Compared with the corresponding evolution algorithms, such as orthogonal genetic algorithm with quantization (OGA/Q), the simulation results of several benchmark functions with high dimension show that DSWOA can acquire satisfied solution, has also a better stability, and a fast convergence rate. Therefore, it is feasible to solve high-dimensional optimization problems.
Keywords :
function approximation; genetic algorithms; social sciences; Milgram sociological experiment; OGA-Q; decimal coding small world optimization algorithm; high dimensional function optimization; imitated society; orthogonal genetic algorithm with quantization; satisfactory solution; small world algorithm; Algorithm design and analysis; Biological information theory; Biological system modeling; Design optimization; Evolutionary computation; Geography; Mechanical engineering; Process design; Simulated annealing; Social network services;
Conference_Titel :
Computational Intelligence in Robotics and Automation (CIRA), 2009 IEEE International Symposium on
Conference_Location :
Daejeon
Print_ISBN :
978-1-4244-4808-1
Electronic_ISBN :
978-1-4244-4809-8
DOI :
10.1109/CIRA.2009.5423233
