
Course unit
LEARNING DYNAMICAL SYSTEMS
INP7080354, A.A. 2019/20
Information concerning the students who enrolled in A.Y. 2018/19
ECTS: details
Type 
ScientificDisciplinary Sector 
Credits allocated 
Core courses 
INGINF/04 
Automatics 
9.0 
Course unit organization
Period 
First semester 
Year 
2nd Year 
Teaching method 
frontal 
Type of hours 
Credits 
Teaching hours 
Hours of Individual study 
Shifts 
Lecture 
9.0 
72 
153.0 
No turn 
Prerequisites:

The course requires the basic knowledge of: signals and systems, data analysis, systems theory, linear algebra. 
Target skills and knowledge:

At the end of the course the student will have developed the basic knowledge on the following aspects of Systems Engineering:
 Knowledge of the principal methods for learning dynamic systems
 Ability of applying the principal methods for learning dynamic systems
 Ability of using the Matlab toolbox “System Identification” 
Examination methods:

The expected knowledge and abilities are checked through an exam divided in three parts: theory, lab and presentation. In the theoretical part the student takes a written test which contains 3 or 4 questions. A question can regard a proof or an exercise saw in the lectures or a slight modification of them. In the lab part the student must do 5 homeworks, each of them corresponds to a lab lecture. In each homework the student must develop a Matlab code and answer to some questions pertinent to the lab lecture. The last part of the exam is not mandatory and is conducted in groups of two, three or four students. Each group have to study and then present a research paper given by the instructor and pertinent to the topics of the course. 
Assessment criteria:

The abilities and knowledge acquired are verified through the following assessment criteria:
 Completeness of the acquired knowledge
 Exactness of the logical steps to argue an answer
 Correct use of technical terminology
 Correct use of the learned methods in contexts which are different from the ones saw in the course 
Course unit contents:

Fisherian Static Estimation;
Bayesian Static Estimation;
Background on Dynamic Models;
PEM method;
Kernelbased PEM method;
Recursive Methods;
Model Structure Determination 
Planned learning activities and teaching methods:

The learning activities are constituted by traditional lessons which address the contents of the course and by 5 labs in which the different methods saw in the lessons are applied to real and synthetic data. 
Additional notes about suggested reading:

All the material presented in the lectures is available in the moodle platform. 
Textbooks (and optional supplementary readings) 

James G., Witten D., Hastie T., and Tibshirani R., An introduction to statistical learningwith applications in R. New York [etc.]: Springer, 2013.

Rasmussen, Carl Edward; Williams, Christopher K.I., Gaussian processes for machine learningCarl Edward Rasmussen, Christopher K.I. Williams. Cambridge: Mass., MIT Press, 2006.

Papoulis, Athanasios; Pillai, Unnikrishna S., Probability, random variables, and stochastic processes. Boston [etc.]: McGrawHill, 1991.

Chen T., Ohlsson H., Ljung L., On the estimation of transfer functions, regularizations and Gaussian processesRevisited. : Automatica n.48, 15251535, 2012.

Petersen K., Pedersen, M., The matrix cookbook. : Technical University of Denmark, Vol. 7, 2008.

Innovative teaching methods: Teaching and learning strategies
 Laboratory
 Case study
 Working in group
 Questioning
 Loading of files and pages (web pages, Moodle, ...)
 Learning journal
Innovative teaching methods: Software or applications used
 Moodle (files, quizzes, workshops, ...)
 Latex
 Matlab
Sustainable Development Goals (SDGs)

