Restartings: A Technique to Improve Classic Genetic Algorithms' Performance

In this contribution, a way to enhance the performance of the classic Genetic Algorithm is proposed. The idea of restarting a Genetic Algorithm is applied in order to obtain better knowledge of the solution space of the problem. A new operator of "insertion" is introduced so as to exploit (utilize) the information that has already been collected before the restarting procedure. Finally, numerical experiments comparing the performance of the classic Genetic Algorithm and the Genetic Algorithm with restartings, for some well known test functions, are given. Keywords--Genetic Algorithms, Restartings, Search space exploration, Search space exploitation.

• Type: Conference
• Year: 2004
• Where: IJIT
• Authors: Grigorios N. Beligiannis, Georgios A. Tsirogiannis, Panayiotis E. Pintelas
• Keywords: Artificial Intelligence, Classic Genetic Algorithm, Genetic Algorithm, IJIT 2004, Search Space