Building Blocks Identification Based on Subtree Sample Counts for Genetic Programming

Often, the performance of genetic programming (GP) is explained in terms of building blocks -- high-quality solution elements that get gradually assembled into larger and more complex patterns by the evolutionary process. However, the weak theoretical foundations of GP building blocks causes their role in GP evolutionary dynamics to remain still somewhat of a mystery. This paper presents a methodology for identifying GP building blocks based on their respective sample counts in the population (defined as the number of times they were sampled by the recombination operators and included in surviving offspring). Our approach represents a problem-independent way to identify important solution elements based on their influence on the evolutionary process.