Ir arriba

Accuracy improvement of Deep Neural Networks through preprocessing and neural structure tuning techniques. An approach to time-series models.

The project consists in carrying out an extensive research on the different alternatives for accuracy boosting of deep learning models, especially those focused on Neural Networks, such as feature selection algorithms, hybrid architectures and Hyperparameter optimization techniques. The aim is to make a comparison of the performance of each method, as well as to suggest alternatives for improvement.


Mónica López-Tafall Criado