Evolutionary Design of Deep Neural Networks

Artificial Intelligence is the science of combining technological and theoretical concepts from multiple disciplines aimed at creating similar behavior in machines to that found in humans. The present work investigates methods of improving the field of pattern classification by applying evolutionary algorithms to train deep neural networks architectures. A significant advantage of evolutionary algorithms over backpropagation is that the neuroevolutionary algorithms allows for training parts of the deep neural network. The framework developed for the present work can be adapted to the required memory and processing resources of the target hardware platform in different application scenarios.