deep learning

DashAI: a graphical toolbox for training, evaluating and deploying state-of-the-art AI models.

The objective of the course is to introduce students to the area of learning based on deep neural networks, commonly known as Deep Learning. The techniques used in this area have been fundamental in the latest advances in Artificial Intelligence, particularly in tasks such as natural language processing and computer vision. During the course, the student will be exposed to the theory behind Deep Learning models, will understand how they work, their possible uses, and will be able to build and train models that allow them to solve real problems.

The GLUES Benchmark aims to collect different sources of tasks for evaluating Spanish Language Models in a unified fashion in order to develop and allow the growth of the Spanish NLP Community.

The purpose of the Natural Language Processing course is to introduce students to the discipline of natural language processing (NLP). The student will use this method to solve tasks, with delimited solutions regarding their relevance and size. This discipline studies the design of methods and algorithms that receive as input and/or produce as output data in the form of natural language (e.g., text, voice). The course focuses on text processing although applications in speech processing are mentioned.

DWUG ES: Diachronic Word Usage Graphs for Spanish

Repository that gathers a compilation of corpus in Spanish language with a size of 3B words.

The purpose of this course is to introduce students to statistical thinking, a systematic way of thinking to describe “real world” phenomena taking into account their inherent uncertainty and using data for decision making. The course introduces the fundamentals of statistical thinking, which are descriptive data analysis, statistical programming in R, and probability theory. It then teaches the two most important schools of statistical inference: frequentist inference and Bayesian inference. Finally, advanced topics such as model evaluation using information theory, directed graphical models for modeling conditional dependencies between variables, and multilevel models are discussed.