1996 | ReviewPaper | Buchkapitel
Encoding scheme issues for open-ended artificial evolution
verfasst von : Nick Jakobi
Erschienen in: Parallel Problem Solving from Nature — PPSN IV
Verlag: Springer Berlin Heidelberg
Enthalten in: Professional Book Archive
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
This paper examines the ways in which the encoding scheme that governs how phenotypes develop from genotypes may be used to improve the performance of open-ended artificial evolution for design. If an open-ended framework involving variable complexity genetic algorithms is adopted, then the vast majority of the evolutionary effort is spent exploring neutral flat areas of the search space. Domain-specific heuristics may be employed to reduce the time spent on searching these neutral areas, however, and the ways in which domain knowledge may be incorporated into the encoding scheme are examined. Experiments are reported in which different categories of scheme were tested against each other, and conclusions are offered as to the most promising type of encoding scheme for a viable open-ended artificial evolution.