2013 | OriginalPaper | Buchkapitel
Dynamic Fitness Landscape Analysis
verfasst von : Hendrik Richter
Erschienen in: Evolutionary Computation for Dynamic Optimization Problems
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Solving optimization problems with time varying objective functions by methods of evolutionary computation can be grounded on the theoretical framework of dynamic fitness landscapes. In this chapter, we define such dynamic fitness landscapes and discuss their properties. To this end, analyzing tools for measuring topological and dynamical landscape properties are studied. Based on these landscape measures we obtain an approach for drawing conclusion regarding characteristic features of a given optimization problem. This may allow to address the question of how difficult the problem is for an evolutionary search, and what type of algorithm is most likely to solve it successfully. The methodology is illustrated using a well-known example, the moving peaks.