This thesis studied a new variant of Evolutionary Algorithms called Gene Expression Programming. The evolution process of Gene Expression Programming was investigated from the practice to the theory. As a practice level, the original version of Gene Expression Programming was applied to a classification problem and an enhanced version of the algorithm was consequently developed. This allowed the development of a general understanding of each component of the genotype and phenotype separated representation system of the solution employed by the algorithm. Based on such an understanding, a version of the schema theory was developed for Gene Expression Programming. The genetic modifications provided by each genetic operator employed by this algorithm were analysed and a set of theorems predicting the propagation of the schema from one generation to another was developed. Also a set of experiments were performed to test the validity of the developed schema theory obtaining good agreement between the experimental results and the theoretical predictions.

Resumo: Este trabalho apresenta uma metodologia composta de duas fases para resolver problemas de otimização com restrições usando uma estratégia multiobjetivo. Na primeira fase, o esforço concentra-se em encontrar, pelo menos, uma solução factível, descartando completamente a função objetivo. Na segunda fase, aborda-se o problema como biobjetivo, onde se busca a otimização da função objetivo original e maximizar o cumprimento das restrições. Na fase um propõe-se uma estratégia baseada na diminuição progressiva da tolerância de aceitação das restrições complexas para encontrar soluções factíveis. O desempenho do algoritmo é validado através de 11 casos testes bastantes conhecidos na literatura especializada. / Abstract: This work presents a two-phase framework for solving constrained optimization problems using a multi-objective strategy. In the first phase, the objective function is completely disregarded and entire search effort is directed toward finding a single feasible solution. In the second phase, the problem is treated as a bi-objective optimization problem, where the technique converts constrained optimization to a two-objective optimization: one is the original objective function; the other is the degree function violating the constraints. In the first phase a methodology based on progressive decrease of the tolerance of acceptance of complex constrains is proposed in order to find feasible solutions. The approach is tested on 11 well-know benchmark functions. / Orientador: Rubén Augusto Romero Lázaro / Coorientador: José Roberto Sanches Mantovani / Banca: Antonio Padilha Feltrin / Banca: Marcos Julio Rider Flores / Mestre

Made available in DSpace on 2014-06-11T19:22:32Z (GMT). No. of bitstreams: 0 Previous issue date: 2009-02-20Bitstream added on 2014-06-13T19:28:05Z : No. of bitstreams: 1 zini_eoc_me_ilha.pdf: 1142984 bytes, checksum: 4ff93a7fe459a5a56e15da26b7a6dd45 (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Este trabalho apresenta uma metodologia composta de duas fases para resolver problemas de otimização com restrições usando uma estratégia multiobjetivo. Na primeira fase, o esforço concentra-se em encontrar, pelo menos, uma solução factível, descartando completamente a função objetivo. Na segunda fase, aborda-se o problema como biobjetivo, onde se busca a otimização da função objetivo original e maximizar o cumprimento das restrições. Na fase um propõe-se uma estratégia baseada na diminuição progressiva da tolerância de aceitação das restrições complexas para encontrar soluções factíveis. O desempenho do algoritmo é validado através de 11 casos testes bastantes conhecidos na literatura especializada. / This work presents a two-phase framework for solving constrained optimization problems using a multi-objective strategy. In the first phase, the objective function is completely disregarded and entire search effort is directed toward finding a single feasible solution. In the second phase, the problem is treated as a bi-objective optimization problem, where the technique converts constrained optimization to a two-objective optimization: one is the original objective function; the other is the degree function violating the constraints. In the first phase a methodology based on progressive decrease of the tolerance of acceptance of complex constrains is proposed in order to find feasible solutions. The approach is tested on 11 well-know benchmark functions.