Semantics Based Substituting Technique for Reducing Code Bloat in Genetic Programming

Abstract

Genetic Programming (GP) is a technique that allows computer programs encoded as a set of tree structures to be evolved using an evolutionary algorithm. In GP, code bloat is a common phenomenon characterized by the size of individuals gradually increasing during the evolution. This phenomenon has a negative impact on GP performance in solving problems. In order to address this problem, we have recently introduced a code bloat control method based on semantics: Substituting a subtree with an Approximate Terminal (SAT-GP). In this paper, we propose an extension of SAT-GP, namely Substituting a subtree with an Approximate Subprogram (SAS-GP). We tested this method with different GP parameter settings on a real-world time series forecasting problem. The experimental results demonstrate the benefit of the proposed method in reducing the code bloat phenomenon and improving GP performance. Particularly, SAS-GP often achieves the best performance compared to other tested GP systems using four popular performance metrics in GP.