Predictive models for the breeder genetic algorithm i. continuous parameter optimization
Evolutionary Computation
Immune clonal selection algorithm for multiuser detection in DS-CDMA systems
AI'04 Proceedings of the 17th Australian joint conference on Advances in Artificial Intelligence
Learning and optimization using the clonal selection principle
IEEE Transactions on Evolutionary Computation
| Hi-index | 0.00 |
A novel artificial immune system algorithm, Immune Forgetting Clonal Programming Algorithm (IFCPA), is put forward. The essential of the clonal selection inspired operations is producing a variation population around the antibodies according to their affinities, and then the searching area is enlarged by uniting the global and local search. With the help of immune forgetting inspired operations, the new algorithm abstract certain antibodies to a forgetting unit, and the antibodies of clonal forgetting unit do not participate in the successive immune operations. Decimal coding with limited digits makes IFCPA more convenient than other binary-coded clonal selection algorithms in large parameter optimization problems. Special mutation and recombination methods are adopted in the antibody population's evolution process of IFCPA in order to reflect the process of biological antibody gene operations more vividly. Compared with some other Evolutionary Programming algorithms such as Breeder Genetic Algorithm, IFCPA is shown to be an evolutionary strategy which has the ability for solving complex large parameter optimization problems, such as high-dimensional Function Optimizations, and has a higher convergence speed.