Evolutionary Demographic Algorithms
This work addresses resource-intensive genetic algorithm problems for researchers and practitioners, but it is incremental as it builds on existing coarse-grain approaches with specific implementation improvements.
The authors tackled the problem of high resource demands in genetic algorithms by developing a Java-JINI distributed library with sub-populations and a graphical interface, showing that this model delays convergence, maintains higher genetic diversity, and allows more evaluations by distributing them across networked computers compared to traditional methods.
Most of the problems in genetic algorithms are very complex and demand a large amount of resources that current technology can not offer. Our purpose was to develop a Java-JINI distributed library that implements Genetic Algorithms with sub-populations (coarse grain) and a graphical interface in order to configure and follow the evolution of the search. The sub-populations are simulated/evaluated in personal computers connected trough a network, keeping in mind different models of sub-populations, migration policies and network topologies. We show that this model delays the convergence of the population keeping a higher level of genetic diversity and allows a much greater number of evaluations since they are distributed among several computers compared with the traditional Genetic Algorithms.