RT info:eu-repo/semantics/bachelorThesis
T1 Redes neuronales convolucionales
A1 Martín de Diego, Elena
A2 Universidad de Valladolid. Facultad de Ciencias
K1 Aprendizaje automático
K1 Aprendizaje profundo
K1 Redes neuronales artificiales
AB Artificial neural networks are a class of machine learning algorithms involved in many of the mostspectacular applications in Artificial Intelligence. Deep neural networks or multilayer networksproduce the best empirical results in the classification of images or texts. From a theoretical pointof view, understanding the reasons for the success of these algorithms is still a pending issue. Inaddition to the convergence problems of the learning algorithms, the enormous overparameterizationof many types of neural networks makes, perhaps, very likely that the classification rules obtainedfollowing this method suffer from overfitting. The goal of this project is to study the design ofneural networks adapted to image analysis, without abundant parametrization, but oriented totake advantage of the special structure of this type of data. The gain of this approach in terms ofcontrol of the overfit will be studied and applied to the classification of some appropriate imagedataset.
YR 2022
FD 2022
LK https://uvadoc.uva.es/handle/10324/58225
UL https://uvadoc.uva.es/handle/10324/58225
LA spa
NO Departamento de Estadística e Investigación Operativa
DS UVaDOC
RD 14-mar-2025