External Concepts Reuse in Genetic Programming

Gregory Seront

In this paper we show how concepts synthesised by Genetic Programming to solve a problem can be reused to solve other ones. This aim is achieved by the creation of a concepts library. These concepts can then be injected in a new population in order to solve a problem that needs them. We explore the performance of this approach against the case where the search starts from a totally random population.


This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.