GEP Book   Home   News   Author   Q&A   Tutorials   Downloads   GEP Biblio   Contacts   Visit Gepsoft

C. FERREIRA In J. M. Benitez, O. Cordon, F. Hoffmann, and R. Roy, eds., Advances in Soft Computing: Engineering Design and Manufacturing, pages 257-266, Springer-Verlag, 2003. Function Finding and the Creation of Numerical Constants in Gene Expression Programming Genetic Algorithms with Tree Representations   All genetic algorithms use populations of individuals, select individuals according to fitness, and introduce genetic variation using one or more genetic operators (see, e.g., Mitchell 1996). In recent years different systems have been developed so that this powerful algorithm inspired in natural evolution could be applied to a wide spectrum of problem domains (see, e.g., Mitchell 1996 for a review of recent work on genetic algorithms and Banzhaf et al. 1998 for a review of recent work on GP). Structurally, genetic algorithms can be subdivided in three fundamental groups: i) Genetic algorithms with individuals consisting of linear chromosomes of fixed length devoid of complex expression. In these systems, replicators (chromosomes) survive by virtue of their own properties. The algorithm invented by Holland (1975) belongs to this group and is known as genetic algorithm or GA; ii) Genetic algorithms with individuals consisting of ramified structures of different sizes and shapes and, therefore, capable of assuming a richer number of functionalities. In these systems, replicators (ramified structures) also survive by virtue of their own properties. The algorithm invented by Cramer (1985) and later developed by Koza (1992) belongs to this group and is known as genetic programming or GP; iii) Genetic algorithms with individuals encoded in linear chromosomes of fixed length which are afterwards expressed as ramified structures of different sizes and shapes. In these systems, replicators (chromosomes) survive by virtue of causal effects on the phenotype (ramified structures). The algorithm invented by myself (Ferreira 2001) belongs to this group and is known as gene expression programming or GEP. GEP shares with GP the same kind of ramified structure and, therefore, can be applied to the same problem domains. However, the logistics of both systems differ significantly and the existence of a real genotype in GEP allows the unprecedented manipulation and exploration of more complex systems. Below are briefly highlighted some of the differences between GEP and GP. Home | Contents | Previous | Next