angle-left MULTIOBJECTIVE MEMETIC ALGORITHMS FOR MULTI-TASK SYMBOLIC REGRESSION
Short Description:

This project aims at developing the new generation of symbolic regression methods using a yet unexplored way to represent mathematical functions. We will use memetic algorithms to create mathematical models for symbolic regression. Our memetic computing approach will be data-driven and will use multi-objective optimization and multi-task evolutionary computation for symbolic regression, addressing a core need of many areas of science and technology. A large number of datasets will be investigated to benchmark the new methods.
The expected outcomes will help support our national priorities with new data analytic capabilities. With a strong and interdisciplinary team in three continents, the project will attract international collaboration.
Funded by the Australian Research Council.
Category: Discovery Project.
Field of Research: Artificial Intelligence and Image Processing.