The main objective of this course is to give students some solid knowledge into the tech-niques of pattern recognition and optimization techniques, so will serve as support an appli-cation to a wide range of scientific disciplines and techniques.
More specifically, the skills aimed to develop among the students of the subject can be de-scribed as follows:
1. Apply the techniques of automatic classification and inference for decision making, in-formation extraction and design of complex systems.
2. Draw conclusions from experimental data, whatever the field in which the researcher works.
3. Optimize classifiers, being of interest to highlight the relationship between the choice of component density functions, the number of parameters needed so as to estimate what impels that choice and the amount of data available for a task, relevant feature selection and dimension reduction of experimental vectors.
4. Critically assess the performance of systems and select the best method of classifica-tion and learning of their experimental data.
5. Apply optimization techniques based on stochastic, heuristic and evolutionary meth-ods.
6. Integrate the knowledge from different sources optimally into management according to the incomplete information available: system status, temporal context, multimodal and personal.
The list of topics of the course deals mainly with contents related to machine learning, according to the following plan. Topic
Introduction and methodology. 6h
Bayes decision theory. 6h
Parametric estimation. 6h
Nonparametric estimation. 6h
Pre-processing and feature selection. 4h
Unsupervised Learning. 4h
SVM and CART. 2h
BN, ART and evolutionary methods. 6h
Optimization methods. 2h
Submission of papers. 2h
Classes are using slides with explanations. At the end of the course, the students will present theirs works.
Students complete the course with a final individual character to be publicly submitted as part of efforts to acquire transferable skills of documentation, communication and publication.
The report must be in the typical format for IEEE conference papers (http://www.ieee.org/conferences_events/conferences/publishing/templates  ....) to foster in students not only reading and interpretation of scientific and technical documents, but also the correct wording.
This final work must be eminently practical, and it should apply the techniques described in the course, preferably, to a problem that may be related to the activity of the student or professional researcher.
The final paper will constitute 70% of the final grade. There will be a written exam, which represents 30% of the final grade.