Buy the Book   Home   News   Author   Q&A   Tutorials   Downloads   GEP Biblio   Contacts   Visit Gepsoft

© C. FERREIRA, 2002 (Terms of Use) ISBN: 9729589054 Gene Expression Programming: Mathematical Modeling by an Artificial Intelligence Karva language: The language of GEP   We have seen that each gene codes for a particular sub-ET, and that each sub-ET corresponds to a specific K-expression or open reading frame. Due to the simplicity and elegance of this correspondence, K-expressions are, per se, extremely compact, intelligible computer programs. We have seen how multi-subunit ETs can be easily converted into linear K-expressions, and this can be easily done for any algebraic, Boolean, or non-conventional expression. Indeed, the language of GEP – Karva language – is a versatile representation that can be used to evolve relatively complex programs as simple, extremely compact, symbolic strings. In fact, there is already commercially available software such as Automatic Problem Solver by Gepsoft which automatically converts K-expressions and GEP chromosomes into conventional C++ or Visual Basic functions. Another advantage of Karva notation is that it can be used to evolve complex programs using any programming language. Indeed, the original GEP implementation was written in C++, but it can be done in virtually any programming language. As a comparison, it is worth mentioning that early GP implementations relied greatly on LISP because it is simple to do the sub-tree swapping that occurs during reproduction with this language. In the next chapter, the implementation details of the gene expression algorithm will be fully analyzed, starting with creating an initial population and finishing with selection and reproduction. Home | Contents | Previous | Next