Pareto, ε-Lexicase, and Down-Sampled Tournament: A Comparative Analysis in Genetic Programming

Application process

A completed online application must be submitted by 4.30 pm 22 September 2025. Late or incomplete applications will not be accepted. Any required supporting documentation (including references) must also be received by 4.30 pm on the closing date in order for the application to be considered.

Project number

133

Project description

This project will explore how different selection methods in Genetic Programming (GP)—a technique that automatically generates computer programs to solve problems—can enhance the development of accurate, efficient, and understandable AI models. Specifically, it will compare Pareto tournament selection, ε-lexicase selection, and down-sampled tournament selection to determine which best balances solution quality, computational cost, and model simplicity across various datasets and conditions.

The successful applicant will work with leading researchers in the field: Professor Bing Xue, Dr. Qi Chen, and Professor Mengjie Zhang.

The expected outcomes are a comprehensive analysis of the operators, a set of program codes for experiments, and a final report or conference paper suitable for publication.

This project is for a single student.