The Parallel Genetic Cellular Automata: Application to Global Function Optimization
This work describes massively parallel genetic algorithms inspired by cellular automata models and by large, spatially distributed populations of individuals, as suggested by biological analogies. Models with strict locality and a variety of pseudo-diffusion models are presented. The models are applied to the the global optimization problem of multiextremal multimodal functions. They are tested on a suite of hard standard test functions. Results are then discussed for the various models taking into account the unusual population sizes, their diversity and the role of individual’s migration.
KeywordsGenetic Algorithm Cellular Automaton Cellular Automaton Cellular Automaton Model Parallel Genetic Algorithm
Unable to display preview. Download preview PDF.
- T. Toffoli and N. Margolous, ‘Cellular Automata Machines’, MIT Press, Cambridge MA, 1989.Google Scholar
- P. Spiessens and B. Manderick, ‘A Genetic Algorithm for Massively Parallel Computers’, in ‘Parallel Processing in Neural Systems and Computers’, North-Holland, 1990.Google Scholar
- ‘Connection Machine CM-200 Series Technical Summary’, Thinking Machine Corporation, Cambridge, MA, 1991.Google Scholar
- A. Torn and A. Zilinskas, ‘Global Optimization’, Springer-Verlag, 1989.Google Scholar