Hierarchical Module Generation in Genetic Programming and Its Incremental Evolution

Abstract

This paper presents a new approach to acquire effective modules in Genetic Programming (GP). Tree-structured chromosome can be handled with GP. The performance of GPs highly depends on the design of primitive functions. The aim of this paper is to propose a new method to determine the effective function nodes as modules.
The modules are generated and stored automatically. The obtained modules have a hierarchical structure.
Logic circuit generations are selected to show the feasibility of the proposed method. They can be represented by the minimal complete set, and it is easy to evaluate the effectiveness of the generated modules. Simulation results show the effectiveness of the proposed method. The experiments for the incremental evolution also show the feasibility for the scaling problem in GPs. The results show that the proposed method can generate effective generic modules.