¡¡Chinese Journal of Computers   Full Text
  TitleM-GEP: A New Evolution Algorithm Based on Multi-Layer Chromosomes Gene Expression Programming
  AuthorsPENG Jing1),2) TANG Chang-Jie1) LI Chuan1) HU Jian-Jun1)
  Address1)(Institute of Database and Knowledge Engineering, College of Computer Science, Sichuan University, Chengdu 610065)
2)(Department of Science and Technology, Chengdu Public Security Bureau, Chengdu 610017)
  Year2005
  IssueNo.9(1459¡ª1466)
  Abstract &
  Background 		This paper proposes a new evolution algorithm, M-GEP, based on the new concept of the multi-layer chromosomes in gene expression programming. The algorithm is efficient in the real applications, such as function discovery, electronic circuit evolution, etc. The main contributions include proposing algorithm M-GEP which is based on multi-layer chromosomes, establishing Level-call model and storage structure between the different chromosomes£¬ and suggesting and implementing chromosomes reorganization operator and genes random reorganization operator. Extensive experiments on the traditional single gene and multi-genes GEP show that the average number of generations of M-GEP is reduce to 29%¡«81%.

keywords multi-layer chromosome; M-GEP; heredity evolution; gene expression programming

background This research belong to the National Science Foundation of China project, ¡°The Research on Key Techniques in Knowledge Discovery Based on Gene Expression Programming¡± (60073046), and is also supported by Sichuan Major Science and Technology Project, ¡°The Outliner Floating Population Analyses Based on Knowledge Discovery ¡±(04SG1640). The projects are mainly focus on research of key techniques based on Gene Expression Programming (GEP). GEP is a new technique in knowledge discovering which borrow idea from the creature evolution rules of gene expression. Comparing the traditional evolution algorithm: GA (Genetic Algorithm) and GP(Genetic Programming) with GEP, authors present following key points: (1) GEP combines the advantage of GA and GP, GEP is 100-60000 times fast than GA or GP. (2) GEP has an inborn advantage to weaken the combination explosion. (3) Gene expression programming does not need the hypothesis, but trusts ¡®truth is inside the training data¡¯.
This paper proposes a new evolution algorithm: M-GEP, which is based on the new concept of the multi-layer chromosome in GEP. The algorithm is efficient in the real applications, such as function discovery, electronic circuit evolution, etc.