Title :
A new hybrid method combining genetic algorithm and coordinate search method
Author :
Qiang Long ; Junjian Huang
Author_Institution :
Sch. of Sci., Inf. Technol. & Eng., Univ. of Ballarat, Ballarat, VIC, Australia
Abstract :
This paper proposed a new hybrid method combining genetic algorithm(GA) and coordinate search method (CSM). Genetic algorithm is good at global exploration but bad at accuracy and local search. Whereas, coordinate search method is good at local exploitation, and its accuracy is reliable when searching in a local area. Thus we combine those two methods in this paper to design a hybrid method called genetic algorithm with coordinate search (GACS). Experimental tests shows that this method are good at both global search and local accuracy.
Keywords :
genetic algorithms; search problems; GACS; genetic algorithm with coordinate search hybrid method; global exploration; local exploitation; local search; Conferences;
Conference_Titel :
Advanced Computational Intelligence (ICACI), 2012 IEEE Fifth International Conference on
Conference_Location :
Nanjing
Print_ISBN :
978-1-4673-1743-6
DOI :
10.1109/ICACI.2012.6463337