Evolutionary Intelligent Agents: A Pattern-based Design

J.G. Fernández García de la Rocha, E. Mosqueira-Rey, and V. Moret-Bonillo (Spain)


Intelligent Agents, Evolutionary Computation, DesignPatterns


Intelligent agents must decide, for each situation within their environment, which action they should perform in order to best achieve their objectives. In other words, they must learn a strategy or a policy which will allow them to choose the best actions for their particular group of goals. Although generally a slow process, it can be speeded up through the use of evolutionary algorithms. This study describes Evolutionary intelligent agents, a hybrid system based on the combination of these two techniques. These agents are capable of adapting to their environment by means of evolutionary learning. To design this architecture for these agents, we have used two classical object-oriented tools UML modelling and design patterns.

Important Links:

Go Back