Evolutionary Search for Feature Selection
This project was designed to formalize an evolutionary search based feature selection methodology for use specifically with the ABCD dataset. A key trait of this project was it was designed to run within a high performance computing environment, with computations spread across multiple compute nodes.
The evolutionary search was also designed with built-in multi-objective age pareto front optimization.
Checkout the project github for more information.