Towards computing the parameters of the simple genetic algorithm
Publikation/Tidskrift/Serie: Proceedings of the 2001 Congress on Evolutionary Computation
Förlag: IEEE--Institute of Electrical and Electronics Engineers Inc.
The problem of finding appropriate probabilities for crossover and mutation with respect to resampling may be addressed using the Markov chain model. Our efforts in this direction lead through a simplification of the mixing matrix incorporating both probabilities. We present the simplification and discuss some of its ramifications. We expect that it may lead to some improvement of the computational properties of the Markov chain model of the simple genetic algorithm.
- Computer Science
- mutation probability
- genetic algorithms
- crossover probability
- Markov processes
- mixing matrix
- Markov chain model
- simple genetic algorithm
CEC2001, Congress on Evolutionary Computation
- ISBN: 0-7803-6657-3