ECTS credits ECTS credits: 6
ECTS Hours Rules/Memories Hours of tutorials: 1 Expository Class: 30 Interactive Classroom: 20 Total: 51
Use languages Spanish, Galician
Type: Ordinary Degree Subject RD 1393/2007 - 822/2021
Departments: Electronics and Computing
Areas: Computer Science and Artificial Intelligence
Center Higher Technical Engineering School
Call: Second Semester
Teaching: With teaching
Enrolment: Enrollable
The subject deals with some of the most important formal paradigms for the treatment and quantification of uncertainty in reasoning. Methods of graphical representation that simplify the analysis of any probabilistic model will be discussed. The subject shows its applicability with multiple examples from science and engineering. The subsequent introduction of decision theory, in combination with probability theory, makes it possible to choose the optimal alternative from the available information, whether incomplete or ambiguous.
Once the student has passed the course:
• You will be familiar with the most important formal paradigms for the treatment and quantification of uncertainty in reasoning.
• You will be able to apply graphical models and Bayesian networks, knowing exact and approximate inference.
• You will be familiar with probabilistic models when solving problems involving uncertainty.
• You will be familiar with decision theory and game theory in problem solving.
Unit 1. Fuzzy Reasoning
1.1 Quantifiers
1.2 Computing with words and perceptions
Unit 2. Probabilistic Reasoning
2.1 Graphical models
2.2 Bayesian networks
2.3 Exact and approximate inference in graphical models
2.4 Learning probabilistic models
Unit 3. Temporal Reasoning
3.1 Fuzzy models
3.2 Markov models
3.3 Kalman filters
3.4 Sequential models
Unit 4. Toma de Decisiones
4.1 Decision theory
4.2 Decision networks
4.3 Complex Decisions
4.4 Game theory
Basic Bibliography
- S. Russell, P. Norvig. Artificial Intelligence. A Modern Approach. 4th ed. Pearson, 2022.
- K.B. Korb, A.E. Nicholson, Bayesian Artificial Intelligence. 2nd Ed. Chapman&Hall/CRC, 2011.
- Daphne Koller and Nir Friedman, Probabilistic Graphical Models: Principles and Techniques, MIT Press, 2009.
- M. J. Kochenderfer, T. A. Wheeler, K. H. Wray. Algorithms for Decision Making, MIT Press, 2022.
Complementary Bibliography
- R. Marín, J.T. Palma (Eds.) Inteligencia Artificial y Sistemas Inteligentes. Ed. McGraw-Hill, 2008.
- A. Darwiche, Modelling and reasoning with Bayesian networks. Cambridge Univ. Press, 2009.
- Pearl, J., Probabilistic Reasoning in Intelligent Systems: networks of plausible inference. Morgan-Kaufmann, 1988.
- Unsupervised learning and inference of Hidden Markov Models in Python (HMMLEARN), https://hmmlearn.readthedocs.io/en/latest/
- Roger R. Labbe Jr, Kalman and Bayesian Filters in Python [Chapters 4 - 8], 2020, https://github.com/rlabbe/Kalman-and-Bayesian-Filters-in-Python
- E. Trillas, L. Eciolaza, Fuzzy Logic. An Introductory Course for Engineering Students, Springer, 2015, ISBN: 978-3-319-14203-6, https://link.springer.com/book/10.1007/978-3-319-14203-6
- Course on Artificial Intelligence. Fernando Sancho Caparrini. http://www.cs.us.es/~fsancho
Competences
BASICS and GENERAL
- CB4] Students must be able to transmit information, ideas, problems and solutions to both specialised and non-specialised audiences.
- GC2] Ability to solve problems with initiative, decision-making, autonomy and creativity.
- GC4] Ability to select and justify the appropriate methods and techniques to solve a specific problem, or to develop and propose new methods based on artificial intelligence.
SPECIFIC
- CE13] Ability to model and design systems based on knowledge representation and logical or approximate reasoning and apply them to different domains and problems, also in contexts of uncertainty.
CROSS-CUTTING
- TR2] Ability to work in a team, in interdisciplinary environments and managing conflicts.
- TR3] Ability to create new models and solutions autonomously and creatively, adapting to new situations. Initiative and entrepreneurial spirit.
Learning Outcomes
- To know the most important formal paradigms for the treatment and quantification of uncertainty in reasoning.
- Know how to apply graphical models and Bayesian networks, knowing exact and approximate inference.
- To be familiar with probabilistic models when solving problems involving uncertainty.
- Knowledge of decision theory and game theory in problem solving.
The teaching methodology will be based on individual work -although sometimes in groups-, discussion with the teacher in class and individual tutorials.
For each topic or thematic block of the lectures, the teacher will prepare the contents, explain the objectives of the topic to the students in class, suggest bibliography, provide them with additional work material, etc. In the lectures, students will work on the competences CB4, CG2, CG4 and CE13. In addition, the lecturers will propose a set of activities to be carried out individually or in groups (assignments, presentations, readings, practicals, etc.), with the aim of facilitating learning. Some of these activities will be assessable and will therefore be compulsory as indicated in the learning assessment system. These activities will allow the development of the previous competences and additionally TR2 and TR3.
Students will work individually or in small groups, with the constant support of the teaching staff. Scripts will be available for practicals, seminars and other activities to be carried out individually or in small groups.
Teaching will be supported by the USC virtual platform in the following way: repository of documentation related to the subject (texts, presentations, recommended readings...) and virtual tutoring of students (e-mail, forums).
The learning assessment considers both the theoretical part (60%) and the practical part (40%). In order to pass the subject, an overall mark of 5 or more out of a maximum of 10 points must be obtained, in accordance with the following criteria:
- Theoretical part (60%): this will be assessed in a single exam to be taken on the official date. The grade of the exam must be equal to or higher than 4 out of a maximum of 10 points in order to pass the whole subject. Otherwise, it will have to be repeated at the make-up exam.
- Practical part (40%): assessment of all compulsory practical activities proposed by the lecturers, according to the following schedule:
• P1: Fuzzy reasoning (deliverable after interactive session 1; weight 10% of the practical part)
• P2: Probabilistic reasoning (delivered after interactive sessions 5 and 6; weight 40% of the practical part).
• P3: Temporal reasoning (to be delivered after interactive sessions 7 and 9; weight 40% of the practical part).
• P4: Decision networks (deliverable after interactive session 10; weight 40% of the practical part)
The assessment of the deliverables may include the completion of a self-assessment questionnaire and/or a face-to-face presentation and discussion session. These assessment activities will be compulsory and may be carried out in interactive class, so that, for the purposes of the provisions of Art.1 of the "Regulamento de asistencia a clase nas ensinanzas oficiais de grao e máster da Universidade de Santiago de Compostela (25/11/2024)", attendance at the sessions where these activities are scheduled will be mandatory, being a requirement the completion of the same which, if not met, will result in a grade of 0.0 in the corresponding deliverable. Except as indicated in this section, class attendance will not have any other assessment in the evaluation system, although attendance at the different teaching activities helps to improve understanding of the subject and the acquisition of competences.
The grade for this part must be equal to or higher than 4 out of a maximum of 10 points in order to pass the whole subject. Submissions with a grade lower than 3 points must be assessed at the second opportunity.
The final grade for the subject will be the arithmetic mean weighted by the percentages indicated above for the theoretical and practical parts, unless the established minimum thresholds are not reached in any assessment item. In the event that the minimum mark required to pass the subject overall is not achieved in one or more parts, the final mark for the opportunity will be the minimum of the marks obtained in those parts.
Students who have neither sat for the examination nor undergone the assessment of any other compulsory activity will be marked as failed.
In order to pass the subject at the second opportunity, students must undergo the assessment of all those compulsory parts pending, in accordance with the above. For the rest, the grades obtained during the course will be retained. Repeating students must follow the same evaluation system as the rest of the students.
In the case of fraudulent performance of exercises or tests, the regulations on the evaluation of the academic performance of students and the review of grades will apply. In application of the ETSE regulations on plagiarism (approved by the Xunta da ETSE on 19/12/2019), the total or partial copying of any practical or theory exercise will result in the failure of the two opportunities of the course, with a grade of 0.0 in both cases.
Classroom work time: 51 hours in total, divided into 30h (lectures), 20h (seminars and practicals), 1h (tutorials).
Personal work time: 99h (total), divided into 39h (self-study of theory and practice) and 60h (assignments, projects and other activities).
It is recommended that students solve, implement, verify and validate all the proposed exercises and practices (not only the assessable ones). It is also considered important to make intensive use of tutorials to resolve any doubts.
It is recommended not to take the course without having previously passed the subjects Programming I, Calculus and Numerical Analysis, Programming II, Statistics, Mathematical Optimisation, "Algorithms," "Fundamentals of Machine Learning", "Knowledge Representation and Reasoning" and "Basic Algorithms of Artificial Intelligence".
The teaching language is Spanish and Galician, but in the bibliography and in the supplementary material there may be part of the contents in English.
Alberto Jose Bugarin Diz
Coordinador/a- Department
- Electronics and Computing
- Area
- Computer Science and Artificial Intelligence
- Phone
- 881816440
- alberto.bugarin.diz [at] usc.es
- Category
- Professor: University Professor
Jose Maria Alonso Moral
- Department
- Electronics and Computing
- Area
- Computer Science and Artificial Intelligence
- Phone
- 881816432
- josemaria.alonso.moral [at] usc.es
- Category
- Professor: University Lecturer
Ainhoa Vivel Couso
- Department
- Electronics and Computing
- Area
- Computer Science and Artificial Intelligence
- ainhoa.vivel.couso [at] usc.es
- Category
- Xunta Pre-doctoral Contract
| Tuesday | |||
|---|---|---|---|
| 09:00-10:00 | Grupo /CLE_01 | Spanish | IA.01 |
| Wednesday | |||
| 09:00-10:00 | Grupo /CLE_01 | Spanish | IA.01 |
| Thursday | |||
| 09:00-10:00 | Grupo /CLE_01 | Spanish | IA.01 |
| Friday | |||
| 10:00-12:00 | Grupo /CLIL_02 | Spanish | IA.02 |
| 05.26.2026 16:00-20:00 | Grupo /CLE_01 | IA.01 |
| 05.26.2026 16:00-20:00 | Grupo /CLIL_01 | IA.01 |
| 05.26.2026 16:00-20:00 | Grupo /CLIL_02 | IA.01 |
| 05.26.2026 16:00-20:00 | Grupo /CLE_01 | IA.02 |
| 05.26.2026 16:00-20:00 | Grupo /CLIL_01 | IA.02 |
| 05.26.2026 16:00-20:00 | Grupo /CLIL_02 | IA.02 |
| 07.03.2026 16:00-20:30 | Grupo /CLIL_01 | IA.01 |
| 07.03.2026 16:00-20:30 | Grupo /CLIL_02 | IA.01 |
| 07.03.2026 16:00-20:30 | Grupo /CLE_01 | IA.01 |