Fine grained population diversity analysis for parallel genetic programming

In this paper we describe a formalism for estimating the structural similarity of formulas that are evolved by parallel genetic programming (GP) based identification processes. This similarity measurement can be used for measuring the genetic diversity among GP populations and, in the case of multi-population GP, the genetic diversity among sets of GP populations: The higher the average similarity among solutions becomes, the lower is the genetic diversity. Using this definition of genetic diversity for GP we test several different GP based system identification algorithms for analyzing real world measurements of a BMW diesel engine as well as medical benchmark data taken from the UCI machine learning repository.