In this section:
In this section we are going to analyze a different kind of symbolic regression, one that is used for discovering Boolean functions. Furthermore, we will also learn how to implement user defined functions (UDFs) and automatically defined functions (ADFs) in GEP. Then we will analyze how the algorithm performs with these new tools so that we can not only understand how they work but also fully explore each one of them. The problem we are going to analyze in this section consists of the well known
odd-n-parity problem. And we are going to discover different odd-parity functions using three different approaches: the first consists of the basic GEA; the second consists of the basic GEA with UDFs; and the third consists of the basic GEA with ADFs. Despite the fact that, in the examples provided in this book, UDFs and ADFs are only used in problems of logic synthesis, they can obviously be used in all the range of problems analyzed in this work.
|