home
| Evolutionary
Algorithms appears like a problem solving technique
biologically
inspired (in the evolution principles), they are essentially
characterized by their implementation facility through computational
systems and their wide application range specially in
optimization and machine learning. The essence of this algorithm type is the transformation of an individual set, each represents a possible solution of the problem, through the application of genetic operators, what conform a process guided by the adaptation capacity of the individuals to the environment. This environment is definend by restrictions and own considerations of each problem and, like in natural evolution, the best individuals, those with greater adaptation level, will survive. From its origin in the 60's, the research in Evolutionary Algorithms has constructed a framework, with detractorss and defenders in the scientific community, that allow to approach to the process of problem solving from a different and, in some cases, superior point of view to the classic techniques like in optimization. Actually we encounter several techniques for the application of Evolutionary Algorithms, in wich are the Genetic Algorithms, Evolutionary Estrategies and the Genetic Programming; although each one of these arose under conjunctures and different necessities, they conserve the essence to evolve solutions through genetic operators and selection of the fittest individuals. |
|