ECTS credits ECTS credits: 4.5
ECTS Hours Rules/Memories Hours of tutorials: 1 Expository Class: 10 Interactive Classroom: 30 Total: 41
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: Sin docencia (Extinguida)
Enrolment: No Matriculable
To understand the procedure used by supervised machine learning to predict values using labeled examples, managing the most efficient classification and regression approaches available nowadays. To group data based on similarity among them using clustering methods. To select or transform the data properties keeping their value and meaning using dimensionality reduction techniques.
The subject website includes the whole subject material:
http://persoal.citius.usc.es/manuel.fernandez.delgado/apaut
1. Classificación and regressión: validation methodology, nearest neighbor methods
2. Linear discriminant analysis (LDA) and linear regression
3. Classification and regression trees
4. Artificial neural networks and deep networks
5. Support vector machines (SVM)
6. Ensembles: bagging, boosting and random forest
7. Clustering
8. Dimensionality reduction
9. Reinforcement learning
Basic:
C.M. Bishop. Pattern recognition and machine learning, Springer, 2006, ISBN: 978-0387310732
R.O. Duda, P.E. Hart, D.G. Stork. Pattern classification. Wiley Interscience, 2000. ISBN: 978-0471056690
R.S. Sutton, A.G. Barton. Reinforcement learning: an introduction. MIT Press, 2nd Edition, 2018. ISBN: ISBN: 978-0262039246
Complementary:
H. Daume. A course in machine learning. Autopublicado, 2017
P. Harrington. Machine learning in action. O'Reilly, 2012. ISBN 978-1617290183
J. Hurwitz, D. Kirsch. Machine learning for dummies. John Wiley & Sons, Inc., 2018. ISBN 9781119454953
CG8: Knowledge about basic technologies that allow to learn and develop new methods, with flexibility for adapting to new situations.
CG9: Ability to solve problems with iniciative, decision making, autonomy and creativity. Ability to communicate knowledge, abilities and skills of Computer Engineering.
TR1: Instrumentals: ability of analysis and synthesis, organization and planification. Oral and written communication in Galician, Spanish and English. Ability for information management. Problem solving. Decision making.
FB3: Ability to understand the basic concepts of discrete mathematics, logic, algorithmics and computational complexity, applied to solve problems in Computer Engineering.
RI5: Knowledge, administration and maintenance of systems, services and applications.
This subject uses software libraries that implement methods of learning from data that are able to adapt to new situations (competence CG8). Besides, there are processes of information extraction that require creative decision making and to communicate the information extracted from data (CG9) in oral and written forms, alongside with plannification of learning strategies and validation of the methods (TR1). These methods use concepts of discrete mathematics, and the analysis of their algorithmics and computational complexity are fundamental for their application to large volume data (FB3). Finally, the administration of systems and applications that implement the analyzed methods is also required (RI5).
Blackboard lectures: slides that explain the problems studied and the concepts, mathematical formulation and applications of the studied methods.
Computer laboratory sessions: solving of practical problems of classification, regression, clustering, dimensionality reduction and reinforcement learning using machine learning libraries.
The subject material, including the slice of the blackboard lectures and the scripts for the exercises on the programming laboratory sessions is publicly available on the subject website (http://persoal.citius.usc.es/manuel.fernandez.delgado/apaut) for the three scenarios.
COVID-19 emergency plan:
Scenario 1: face-to-face lectures and laboratory sessions.
Scenario 2: telematic lectures (using Microsoft Teams) and face-to-face laboratory sessions.
Scenario 3: telematic lectures and laboratory sessions (Teams).
Continuous assessment: exams with practical exercises about the studied techniques: 30% of the final mark.
Final exam: questions and practice exercises: 70% of the final mark.
COVID-19 emergency plan:
Scenario 1: face-to-face exams of continuous and final assessment.
Scenario 2: telematic exams of continuous assessment (virtual USC) and face-to-face final exam.
Scenario 3: telematic exams of continuous and final assessment (virtual USC).
In case of exam fraud the "Normativa de avaliación do rendemento académico d@s estudant@s e de revisión de cualificacións". will be applied.
Face-to-face work:
Blackboard lectures: 10h
Computer laboratory sessions: 30h
Assessment exams: 5h
Total: 45h
Personal work:
Autonomous study: 19h
Exercise solving: 7h
Computer programming: 28h
Exam preparation: 13h
Total: 67h
Attendance to the blackboard lectures and computer laboratory sessions, and solving the proposed exercises using the machine learning libraries employed on the laboratory.
The virtual USC will be also used:
http://cv.usc.es
COVID-19 emergency plan: see items "Teaching methodology" and "Assessment system" for the adaptations to scenarios 2 and 3.
Manuel Fernandez Delgado
Coordinador/a- Department
- Electronics and Computing
- Area
- Computer Science and Artificial Intelligence
- Phone
- 881816458
- manuel.fernandez.delgado [at] usc.es
- Category
- Professor: University Lecturer
| Tuesday | |||
|---|---|---|---|
| 16:30-17:30 | Grupo /CLE_01 | Galician | Classroom A4 |
| Thursday | |||
| 09:30-12:00 | Grupo /CLIL_01 | Galician | Lab PP-2 |
| 05.26.2021 16:00-20:45 | Grupo /CLE_01 | PROJECTS |
| 05.26.2021 16:00-20:45 | Grupo /CLIL_01 | PROJECTS |
| 07.07.2021 16:00-20:45 | Grupo /CLIL_01 | Classroom A3 |
| 07.07.2021 16:00-20:45 | Grupo /CLE_01 | Classroom A3 |