Adaptation in natural and artificial systems
Adaptation in natural and artificial systems
Intelligence through simulated evolution: forty years of evolutionary programming
Intelligence through simulated evolution: forty years of evolutionary programming
Genetic Algorithms in Search, Optimization and Machine Learning
Genetic Algorithms in Search, Optimization and Machine Learning
Handbook of Evolutionary Computation
Handbook of Evolutionary Computation
A Fine-Grained Parallel Genetic Algorithm for Distributed Parallel Systems
Proceedings of the 5th International Conference on Genetic Algorithms
A Comparison of Parallel and Sequential Niching Methods
Proceedings of the 6th International Conference on Genetic Algorithms
Building Better Test Functions
Proceedings of the 6th International Conference on Genetic Algorithms
The Distributed Genetic Algorithm Revisited
Proceedings of the 6th International Conference on Genetic Algorithms
Parameter-Free Genetic Algorithm Inspired by ``Disparity Theory of Evolution''
PPSN V Proceedings of the 5th International Conference on Parallel Problem Solving from Nature
An analysis of the behavior of a class of genetic adaptive systems.
An analysis of the behavior of a class of genetic adaptive systems.
Hi-index | 0.00 |
We propose a novel genetic algorithm which we call a parameter-free genetic algorithm (PfGA). The PfGA is inspired by the idea of a biological evolution hypothesis, i.e., the "disparity theory of evolution." The theory is based on different mutation rates in double strands of DNA. Its idea can be extended to a very compact and fast adaptive search algorithm accelerating its evolution based on the variable size of a population and taking a dynamic but delicate balance between exploration (i.e., global search) and exploitation (i.e, local search). The PfGA is not only simple and robust, but it is unnecessary to set almost all the genetic parameters in advance which need to be set up in other genetic algorithms. Furthermore, a uniformly distributed parallel architecture and a master-slave architecture for the PfGA are investigated as an extension. We discuss the performance of the parallel distributed architectures using a general set of function optimization problems including the functions in the first International Contest on Evolutionary Optimization. On the other hand, gene duplication theory was first proposed by a Japanese biologist, Dr. Susumu Ohno, in the 1970's. Inspired by this theory, we develop a gene-duplicating genetic algorithm. Several variants of this algorithm are considered. Individuals with various lengths of genes are evolved based on the PfGA or steady-state GA and then genes with different lengths are concatenated by migrating among subpopulations. To verify the effectiveness of the gene-duplicating genetic algorithm, we also performed a comparative study using the general set of function optimization problems.