Automatic design and optimization of neural network using evolutionary algorithm NSGA2

Using a multi-objective algorithm to design and optimize the structure of a neural network for processing a selected benchmark dataset. A set of networks will be designed that are Pareto optimal in terms of accuracy, computational complexity, and memory usage.