ABSTRACT
This chapter introduces a new approach to Genetic Programming (GP), based on a numerical, i.e., GMDH(Group Method of Data Handling)-based, technique, which integrates a GP-based adaptive search of tree structures, and a local parameter tuning mechanism employing statistical search (i.e., a system identification technique). In traditional GP, recombination can cause frequent disruption of building blocks, or mutation can cause abrupt changes in the semantics. To overcome these difficulties, we supplement traditional GP with a local hill-climbing search, using a parameter tuning procedure. More precisely, we integrate the structural search of traditional GP with a multiple regression analysis method and” establish our adaptive program called “STROGANOFF” (i.e., STructured Representation On Genetic Algorithms for NOnlinear Function Fitting). The fitness evaluation is based on a “Minimum Description Length ” (MDL) criterion, which effectively controls the tree growth in GP. We demonstrate its effectiveness by solving several system identification (numerical) problems and compare the performance of STROGANOFF with traditional GP and another standard technique (i.e., “radial basis functions”). The effectiveness of this numerical approach to GP is demonstrated by successful application to computational finance.
