On the Design of Adaptive Control Strategies for Evolutionary Algorithms

This paper focuses on the design of control strategies for Evolutionary Algorithms. We propose a method to encapsulate multiple parameters, reducing control to only one criterion. This method allows to define generic control strategies independently from both the algorithm’s operators and the problem to be solved. Three strategies are proposed and compared on a classical optimization problem, considering their generality and performance.

• Type: Conference
• Year: 2007
• Where: AE
• Authors: Jorge Maturana, Frédéric Saubion
• Keywords: AE 2007, Artificial Intelligence, Classical Optimization Problem, Control Strategies, Generic Control Strategies