3 Review(s)

Computational Intelligence comprises concepts, paradigms, algorithms, and implementations of systems that are supposed to exhibit intelligent behavior in complex environments. It relies heavily on sub-symbolic, predominantly nature-analog or at least nature-inspired methods. These methods have the advantage that they tolerate incomplete, imprecise and uncertain knowledge and thus also facilitate finding solutions that are approximatively, manageable and robust at the same time. The choice of topics in this book reflects the most important fields in the area of computational intelligence. Classical fields such as artificial neural networks, fuzzy systems, and evolutionary algorithms are described in considerable detail. However, methods such as ant colony optimization and probabilistic graphical models are discussed as well, although a complete coverage of all approaches and developments is clearly impossible to achieve in a single volume. Rather than to strive for completeness, our goal is to give a methodical introduction to the area of Computational Intelligence. Hence we try not only to present fundamental concepts and their implementations, but also explain the theoretical background of proposed problem solutions.