Genetic Programming Theory and Practice: A Fifteen-Year Trajectory

TL;DR

This paper analyzes the GPTP workshop series from 2003 to 2017 to trace the thematic evolution of genetic programming. It documents a shift from a theory–practice split toward widespread adoption of best-practice techniques, with symbolic regression as a dominant yet evolving application area, and growing emphasis on usability, scalability, and real-world impact. It notes the emergence of diverse representations, big data and ML integration, cloud computing, and new operators such as semantic GP and lexicase selection. The analysis highlights recurring challenges: need for domain knowledge, lack of turnkey GP systems, and tension between theory and practice, while also forecasting ongoing progress in open-ended evolution and automatic programming.

Abstract

The GPTP workshop series, which began in 2003, has served over the years as a focal meeting for genetic programming (GP) researchers. As such, we think it provides an excellent source for studying the development of GP over the past fifteen years. We thus present herein a trajectory of the thematic developments in the field of GP.