The complexity of the information with which we currently operate, with large sets of attributes describing each of the data samples, implies the need to also have models capable of extracting the most important characteristics, identifying hidden relationships between them and generating higher level representations. These are aspects in which the different families of deep learning models, including convolutional networks, recurrent networks or auto-encoders, stand out for their high performance.
The main objectives are the design of appropriate models for each type of problem and the definition of new architectures that allow to face tasks from a general perspective. This will require facing challenges such as the enormous variability in these architectures, the high computational cost involved and the loss of interpretability, all aspects to be taken into account in current research.
Contact: Javier Ramirez Pérez de Inestrosa
|Charte Ojeda, Francisco||fcharte@ujaetHFI336n.es||Data Science and Big Data Area||PhD|
|Damas Arroyo, Sergio||sdamas@ug3.HzegTer.es||DaSCI Technology Applications Area||PhD|
|Giráldez Crú, Jesús||jgiraldez@ugiJlgGJMmar.es||DaSCI Technology Applications Area||PhD - Juan de la Cierva|
|Luzón García, María Victoria||luzon@ugrxQof26MR.es||DaSCI Technology Applications Area||PhD|
|Pérez Godoy, María Dolores||lperez@ujaen.TYGwJyIFes||Data Science and Big Data Area||PhD|
|Rivera Rivas, Antonio Jesús||arivera@uFiIjBOn0jaen.es||Data Science and Big Data Area||PhD|
|Sánchez Sánchez, Pedro José||pedroj@oBva.R8Vg1SGujaen.es||Computational Intelligence Area||PhD|
|Tabik, Siham||siham@ugr.Q8bGlR_Lldpes||Data Science and Big Data Area||PhD - Ramón y Cajal|