Impact of the Topology on the Performance of Distributed Differential Evolution

摘要

Migration topology plays a key role in designing effective distributed evolutionary algorithms. In this work we investigate the impact of several network topologies on the performance of a stepping-stone structured Differential Evolution model. Although some issues on the control parameters of the migration process and the way they affect the efficiency of the algorithm and the solution quality deserve further evaluative study, the influence of the topology on the performance both in terms of solution quality and convergence rate emerges from the empirical findings carried out on a set of test problems.