ECTS credits ECTS credits: 6
ECTS Hours Rules/Memories Hours of tutorials: 1 Expository Class: 21 Interactive Classroom: 21 Total: 43
Use languages English
Type: Ordinary subject Master’s Degree RD 1393/2007 - 822/2021
Departments: Electronics and Computing, External department linked to the degrees
Areas: Computer Science and Artificial Intelligence, Área externa M.U en Intelixencia Artificial
Center Higher Technical Engineering School
Call: First Semester
Teaching: With teaching
Enrolment: Enrollable | 1st year (Yes)
The course introduces the three main paradigms in the field of machine learning: supervised, unsupervised and reinforcement learning. The student will be trained in the generation of prediction models (regression and classification), also considering the possibility of combining different techniques to improve performance. Strategies will also be described to optimize the performance of both learning, by means of preprocessing and feature extraction techniques, and of the models generated, by means of their regularization and evaluation.
1. Supervised learning: Introduction to learning, Artificial Neural Networks, Support Vector Machines, Decision trees, Regression, Instance-based learning
2. Ensemble modeling: Basic and advanced ensemble modelling
3. Preprocessing, evaluation and regularization: Data preprocessing. Model creation and evaluation, Complexity and Regularization.
4. Unsupervised learning: Clustering, Unsupervised neural networks
5. Reinforcement learning: Markov decision processes, Reinforcement learning
Basic:
• Ethem Alpaydin (2004). Introduction to Machine Learning. MIT Press
• T.M. Mitchell (1997). Machine Learning. McGraw Hill
• Richard Sutton, Andrew Barto (2018). Reinforcement Learning. Second Edition. MIT Press
Complementary:
• Andrew Webb (2002). Statistical Pattern Recognition. Wiley
• D. Borrajo, J. González, P. Isasi (2006). Aprendizaje automático. Sanz y Torres.
• Basilio Sierra Araujo (2006). Aprendizaje automático: conceptos básicos y avanzados. Aspectos prácticos utilizando el software WEKA. Pearson Education
• Saso Dzeroski, Nada Lavrac (2001). Relational Data Mining. Springer.
• David Aha (1997). Lazy Learning. Kluwer Academics Publishers
BASIC AND GENERAL
GC2 - Successfully tackle all the stages of an Artificial Intelligence project.
GC3 - Search and select useful information needed to solve complex problems, handling with fluency the bibliographic sources of the field.
GC4 - Elaborate adequately and with some originality written compositions or motivated arguments, write plans, work projects, scientific articles and formulate reasonable hypotheses in the field.
CG5 - Work in teams, especially multidisciplinary teams, and be skilled in time management, people and decision making.
CB6 - Possess and understand knowledge that provides a basis or opportunity for originality in the development and/or application of ideas, often in a research context.
CB7 - That students know how to apply acquired knowledge and problem-solving skills in new or unfamiliar environments within broader (or multidisciplinary) contexts related to their area of study.
CB8 - That students are able to integrate knowledge and face the complexity of making judgments based on information that, being incomplete or limited, includes reflections on the social and ethical responsibilities linked to the application of their knowledge and judgments.
CB9 - That students know how to communicate their conclusions and the ultimate knowledge and reasons that support them to specialized and non-specialized audiences in a clear and unambiguous way.
TRANSVERSALS
CT3 - Use the basic tools of information and communication technologies (ICT) necessary for the exercise of their profession and for lifelong learning.
CT4 - Develop for the exercise of a citizenship respectful of democratic culture, human rights and gender perspective.
CT7 - Develop the ability to work in interdisciplinary or transdisciplinary teams, to offer proposals that contribute to sustainable environmental, economic, political and social development.
CT8 - Value the importance of research, innovation and technological development in the socioeconomic and cultural progress of society.
CT9 - Have the ability to manage time and resources: develop plans, prioritize activities, identify critical ones, set deadlines and meet them.
SPECIFIC
CE10 - Ability to build, validate and apply a stochastic model of a real system from observed data and critical analysis of the results obtained.
SC11 - Understanding and mastery of the main techniques and tools of data analysis, both from the statistical point of view and machine learning, including those dedicated to the treatment of large volumes of data, and ability to select the most appropriate for problem solving.
CE12 - Ability to plan, formulate and solve all stages of a data project, including understanding and mastery of basic fundamentals and techniques for searching and filtering information in large data collections.
CE15 - Knowledge of computer tools in the field of machine learning, and ability to select the most appropriate for solving a problem.
The methodology includes the expository method / lecture, laboratory practices, tutorials, problem-based learning. It will be carried out with the following training activities:
1) Problem-based learning : these are sessions whose objective is that students acquire certain skills based on the resolution of exercises, and projects that require the student to apply the knowledge and skills developed during the course. These sessions may require the student to present orally the solution to the problems posed. The work carried out by the students can be done in work groups.
2) Theory classes: Oral exposition complemented with the use of audiovisual media and the introduction of some questions directed to the students, with the purpose of transmitting knowledge and facilitating learning. In addition to the time of oral exposition by the professor, this formative activity requires the student to dedicate some time to prepare and review on their own the materials object of the class.
3) Practical laboratory classes: classes dedicated to the development of practical work involving the resolution of complex problems, and the analysis and design of solutions that constitute a means for their resolution. This activity may require students to present their work orally. The work done by the students can be done in work groups.
A continuous assessment will be carried out as part of the laboratory practice; whereas the summative assessment will cover the implementation of a supervised project and one final exam.
1. Laboratory practice: it focuses on the resolution of problems, by means of an appropriate application of AI techniques explained in theory. The problems will be grouped into four assignments, which should be solved in pairs. The proposed assignments will stimulate critical thinking in the solution of problems. The four assignments are evaluated with a weighting of 20% of the final grade.
2. Supervised projects: students will work in groups (e.g., 4 people) to solve a classification or prediction problem. The project will consist of the implementation and writing a report on the resolution of the selected problem. The report will include a bibliographic review of the most important related works, it must be written in English (documentation on the problem to be solved, methodology used, and comparison of the results found in the application of the different techniques, as well as a critical evaluation of both the results obtained and the information used). The project is evaluated with a weighting of 30% of the final grade.
3. Final exam: test questions about the contents of the course, based on the different machine learning techniques and their applications. The exam will be evaluated with a weighting of 50% of the final grade.
Students must achieve at least 40% of the maximum mark for each part (theory, practice) and in any case the sum of both parts must exceed 5 to pass the subject. If any of the above requirements is not met, the grade of the call will be established according to the lowest grade obtained.
The deliveries of the practices must be made within the period established in the virtual campus and must follow the specifications indicated in the statement both for their presentation and their defense. Students will have the condition of "Presented" if you attend the theoretical test in the official evaluation period.
Second opportunity: The evaluation will be carried out with the same criteria described above, and a new term will be opened for the delivery of the practical works.
In the case of fraudulent completion of exercises or tests, the Regulations for evaluating the academic performance of students and reviewing qualifications will be applied. In application of the corresponding regulations on plagiarism, the total or partial copy of any practice or theory exercise will be evaluated with a grade of 0.
A1: Theory classes: 21 classroom hours, 42 hours total dedication.
A2: Practical laboratory classes: 14 classroom hours, 60 hours total dedication.
A3: Problem-based learning, seminars, case studies and projects: 7 classroom hours, 48 hours total dedication.
Weekly study of the subject is recommended
Nelly Condori Fernandez
Coordinador/a- Department
- Electronics and Computing
- Area
- Computer Science and Artificial Intelligence
- n.condori.fernandez [at] usc.es
- Category
- Professor: LOU (Organic Law for Universities) PhD Assistant Professor
| Tuesday | |||
|---|---|---|---|
| 17:00-18:30 | Grupo /CLIL_01 | English | IA.02 |
| 18:30-20:00 | Grupo /CLE_01 | English | IA.02 |
| 01.13.2025 16:00-20:00 | Grupo /CLIL_01 | IA.02 |
| 01.13.2025 16:00-20:00 | Grupo /CLE_01 | IA.02 |
| 06.19.2025 16:00-20:00 | Grupo /CLIL_01 | IA.02 |
| 06.19.2025 16:00-20:00 | Grupo /CLE_01 | IA.02 |