This course examines the theoretical bases that support the main models used in pattern recognition applications. The emphasis is on learning techniques, both statistical models and Artificial Neural Networks and explains its usefulness in practical problems of signal processing and image processing.
The main goal is teaching students to acquire sufficient skills to deal with an application that provides data representative of an input-output system, natural or artificial, and be able to build a model that explains the system and answer in an analogous way, both as a functional approximation problem and as a classification problem. To achieve this objective, the student should be able to develop proficiency in the use of concepts such as curse of dimensionality, generalisation, sample size, complexity of the model, approximation error, error estimation, empirical error, bias and variance of the model, etc..
