Long-Term Evolution of Genetic Programming Populations
This addresses the long-term evolutionary dynamics in genetic programming, but it is incremental as it focuses on specific observations without broad methodological changes.
The study evolved binary mux-6 trees for up to 100,000 generations, producing programs with over 100 million nodes, and found that the population did not evolve building blocks but suggested a limit to bloat, with periods of 100% functional convergence lasting tens to hundreds of generations.
We evolve binary mux-6 trees for up to 100000 generations evolving some programs with more than a hundred million nodes. Our unbounded Long-Term Evolution Experiment LTEE GP appears not to evolve building blocks but does suggests a limit to bloat. We do see periods of tens even hundreds of generations where the population is 100 percent functionally converged. The distribution of tree sizes is not as predicted by theory.