An Improved Evolutionary Algorithm as Function Optimizer
Sanghamitra Bandyopadhyay, Ujjwal Maulik · IETE Journal of Research · 2000
In this paper an efficient evolutionary search algorithm, which can exploit the enhanced searching capability of parallel genetic algorithms while being sequential in nature, has been proposed. The performance of the algorithm with and without migration of chromosomes has been studied. The necessity of information exchange, in the form of migration between different processes, for good performance of the proposed algorithm is adequately established for a variety of function optimization problems. Variation of performance of the proposed algorithm with the number of migrations is demonstrated experimentally; thereby indicating that a proper choice of the frequency of migration and number of processes are crucial.