| Abstract |
In this talk, we present our current research on the role of explicit niching and communication messages in distributed evolutionary multi-objective optimization. In this work, localization is employed to implement explicit niching. Several options are selected for communication messages including the explicit communication of solutions and the communication of statistical information.
A distributed system using the framework of local models, called LOD-EMOS, is developed to support distributed computing for an evolutionary multi-objective optimization. This system provides flexibility in applying different architectures such as master/slave, island as well as the hybridization of the two.
It is found that communicating statistical information is efficient and reduces the communication overload dramatically. Furthermore, statistical information can be obtained easily by the use of explicit niching during the course of action.
|
