memetic algorithm

Memetic algorithms is a population-based approach for heuristic search in optimization problems. For some problem domains they have been shown to be more efficient than genetic algorithms. Some researchers view them as hybrid genetic algorithms or parallel genetic algorithms.From the view of Genetic Algorithm, if GA is combined with some kinds of Local Search, the algorithm is termed as memetic algorithm.Memetic algorithms are the subject of intense scientific research and have been successfully applied to a multitude of real-world problems ranging from the construction of university exam timetables, to the prediction of protein structures and the design of spacecraft trajectories.
See more at Wikipedia.org...

<algorithm> A genetic algorithm or evolutionary algorithm which includes a non-genetic local search to improve genotypes. The term comes from the Richard Dawkin's term "meme".
One big difference between memes and genes is that memes are processed and possibly improved by the people that hold them - something that cannot happen to genes. It is this advantage that the memetic algorithm has over simple genetic or evolutionary algorithms.
These algorithms are useful in solving complex problems, such as the "Travelling Salesman Problem," which involves finding the shortest path through a large number of nodes, or in creating artificial life to test evolutionary theories.
Memetic algorithms are one kind of metaheuristic.
UNLP memetic algorithms home page.
(07 July 1997)