Chinese Journal of Computers   Full Text
Title	The Strategy for Diversifying Initial Population of Gene Expression Programming
Authors	HU Jian-Jun1),2),3) TANG Chang-Jie1) DUAN Lei1) ZUO Jie1) PENG Jing1),4) YUAN Chang-An1),5)
Address	1)(School of Computer Science, Sichuan University, Chengdu 610065) 2)(Information Science School, Guangdong University of Business Studies, Guangzhou 510320) 3)(School of Computer Science & Engineering, South China University of Technology, Guangzhou 510641) 4)(Department of Science and Technology, Chengdu Public Security Bureau, Chengdu 610017) 5)(Department of Information & Technology, Guangxi Teachers Education University, Nanning 530001)
Year	2007
Issue	No.2(305—310)
Abstract &   Background	Abstract Gene Expression Programming (GEP) is a new genetic algorithm for knowledge discovery. The diversification of initial population is very important to the evolution efficiency and result. In order to produce excellent initial population of GEP, Gene Space Balance Strategy (GSBS) is proposed. The theorem of describing the space of GEP encoding is proved. The simulation experiments show that GSBS can increase the evolutionary efficiency by 20%. The idea of GSBS algorithms can be used in other evolutionary computation else. keywords genetic programming; genetic algorithm; gene expression programming; function mining background This paper is supported by the National Natural Science Foundation of China under grant No.60473071. Gene Expression Programming (GEP) is an effective evolutionary algorithm and is widely used in function mining. The initial population of GEP is an important factor to the evolutionary efficiency. The random way is usually used to produce the initial population. By this way the gene diversity in it is very poor. In order to produce excellent initial population of GEP, Gene Space Balance Strategy (GSBS) is proposed. GSBS increases the diversity in the initial population. The simulation experiments show that GSBS can increase the evolutionary efficiency by 20%. The idea of GSBS algorithm can be used in other evolutionary computation else. This paper is focused on producing excellent initial population of GEP.