Title :
Schema theorem of real-coded nonlinear genetic algorithm
Author :
Cui, Zhi-hua ; Zeng, Jian-chao
Author_Institution :
Division of System Simulation and Computer Application, Taiyuan Heavy Machinery Institute, Shanxi, China
Abstract :
Through the mechanism analysis of simple genetic algorithm (SGA), every genetic operator can be considered as a linear function. So some disadvantages of SGA may be solved if the genetic operators are modified to a nonlinear function. According to the above method, a nonlinear genetic algorithm is introduced. By analysing of the concept of the schema, a new schema theorem is proved. The above work establishes a completely new mathematical analytic method of the genetic algorithms schema theory and contributes much to perfection and development of the theory of genetic algorithm.
Keywords :
genetic algorithms; minimisation; probability; Holland schema; distance space; genetic operator; global minimization; mutation operators; nonlinear crossover; nonlinear genetic algorithm; probability; schema theorem; Algorithm design and analysis; Analytical models; Computational modeling; Computer applications; Computer simulation; Convergence; Genetic algorithms; Genetic mutations; Minimization methods; Performance analysis;
Conference_Titel :
Machine Learning and Cybernetics, 2002. Proceedings. 2002 International Conference on
Print_ISBN :
0-7803-7508-4
DOI :
10.1109/ICMLC.2002.1167442