2012 | OriginalPaper | Chapter
MGPSO – The Managed Evolutionary Optimization
Author : Radosław Z. Ziembiński
Published in: Swarm and Evolutionary Computation
Publisher: Springer Berlin Heidelberg
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
This paper introduces a new modular algorithm MGPSO. It merges random probing, new particle swarm optimization and local search algorithms. The proposed algorithm was implemented according to a new proposal of modular architecture. It allows for flexible mixing different techniques of the optimization in a single optimization. The architecture allows to macro–manage the search process by modifiable set of rules. Thus, a selection of suitable tools for different phases of the optimization depending on current requirements is possible. As a consequence, the modular algorithm achieves good performance acknowledged in performed experiments. The proposed architecture can help in application of machine learning methods for the selection of efficient sequence of tools during the optimization.