Genetic algorithms based on bintree structure encoding
Yongxian Wang · Journal of Tsinghua University(Science and Technology) · 2000
A mathematical description of genetic algorithms is proposed based on the bintree structural representation. Taking polish notation of the functional model as an example, the algorithms' three genetic operators: selection, mutation and crossover, are explained in detail. The convergence of the algorithm is also analyzed. If the space of the bintree structure is limited by the depth of the bintree, the algorithm can be described with Markov chains of finite state space. On the basis of these studies, improved genetic algorithm convergence in probability is proven.