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

C. FERREIRA Invited Tutorial Presented at WSC6, 2001 Gene Expression Programming in Problem Solving Summary   On the one hand, the implementation details of the learning algorithm, gene expression programming, were thoroughly presented, allowing its easy understanding and implementation. On the other hand, the workings of the algorithm were analyzed step-by-step with a simple problem of symbolic regression. Furthermore, the question of constant creation in symbolic regression was discussed comparing two different approaches to solve this problem: one with the explicit use of rational constants, and another without them. The results presented suggest that the latter is best, not only in terms of the accuracy of the evolved models and overall performance evaluated in terms of average best-of-run fitness, but also because the search space is much smaller, reducing greatly the complexity of the system. Moreover, we also saw how GEP efficiently searched for a solution to a complex problem on a five-dimensional parameter space with several extraneous functions, finding an almost perfect solution with an R-square of 0.9999913. Home | Contents | Previous | Next