Migration strategy plays an important role in designing effective distributed evolutionary algorithms. Here, a novel migration model inspired to the phenomenon known as biological invasion is adopted. The migration strategy is implemented through a multistage process involving large invading subpopulations and their competition with native individuals. In this work such a general approach is used within an island parallel model adopting Differential Evolution as the local algorithm. The resulting distributed algorithm is evaluated on a set of well known test functions and its effectiveness is compared against that of a classical distributed Differential Evolution.
A model based on biological invasions for island evolutionary algorithms
De Falco I;Maisto D;Scafuri;Tarantino E
2012
Abstract
Migration strategy plays an important role in designing effective distributed evolutionary algorithms. Here, a novel migration model inspired to the phenomenon known as biological invasion is adopted. The migration strategy is implemented through a multistage process involving large invading subpopulations and their competition with native individuals. In this work such a general approach is used within an island parallel model adopting Differential Evolution as the local algorithm. The resulting distributed algorithm is evaluated on a set of well known test functions and its effectiveness is compared against that of a classical distributed Differential Evolution.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.