%0 Report
%T EASEA : un langage de spécification pour les algorithmes évolutionnaires
%+ Centre de Mathématiques Appliquées (CMAP)
%+ Fractals, complex models and artificial evolution (FRACTALES)
%+ École Nationale Supérieure de Techniques Avancées (ENSTA Paris)
%A Collet, Pierre
%A Schoenauer, Marc
%A Lutton, Evelyne
%A Louchet, Jean
%N RR-4218
%P 17
%I INRIA
%8 2001
%D 2001
%Z Computer Science [cs]/Artificial Intelligence [cs.AI]Reports
%X Evolutionary algorithms are not straightforward to implement and the lack of any specialised language forces users to write their algorithms in C, C++ or JAVA. However, most evolutionary algorithms follow a similar design, and the only really specific part is the code representing the problem to be solved. Therefore, it seems that nothing, in theory, could prevent a user from being able to design and run his evolutionary algorithm from a Graphic User Interface, without any other programming effort than the function to be optimised. Writing such a GUI rapidly poses the problem of saving and reloading the evolutionary algorithm on which the user is working, and translating the information into compilable code. This very much sounds like a specifying language and its compiler. The EASEA software was created on this purpose, and to our knowledge, it is the first and only usable compiler of a language specific to evolutionary algorithms. This reprot describes how EASEA has been designed and the problems which needed to be solved to achieve its implementation.
%G English
%2 https://hal.inria.fr/inria-00000849/document
%2 https://hal.inria.fr/inria-00000849/file/19_RR4218.pdf
%L inria-00000849
%U https://hal.inria.fr/inria-00000849
%~ X
%~ ENSTA
%~ CNRS
%~ INRIA
%~ INRIA-ROCQ
%~ INRIA-RRRT
%~ X-CMAP
%~ X-DEP
%~ X-DEP-MATHA
%~ INRIA_TEST
%~ TESTALAIN1
%~ CMAP
%~ UVSQ
%~ INRIA2
%~ LARA