Publication in the Diário da República: Despacho n.º 13495/2022 - 18/11/2022
10 ECTS; 1º Ano, 1º Semestre, 30,0 PL + 30,0 TP + 30,0 OT , Cód. 390914.
Lecturer
- Gabriel Pereira Pires (1)(2)
(1) Lead Professor
(2) Teaching Professor
Prerequisites
Algebra and statistics and one programming language.
Objectives
The main objective of this course is to provide students with knowledge about machine learning with a focus on regression and supervised classification (classical and deep learning). By the end of this course, it is expected that students will be able to implement all regression and classification steps and apply them to diverse datasets obtained from real problems.
Program
1. Introduction to supervised and unsupervised machine learning;
2. Descriptive statistics;
3. Simple and multiple linear regression. Nonlinear regression. Parameter estimation through Ordinary Least Squares (OLS) and Gradient Descent (GD). Evaluation of regression models;
4. Regularization methods;
5. Normalization methods and dimensionality reduction;
6. Classifiers: Bayes, Linear Discriminant Analysis, Logistic regression, K-Nearest Neighbors, Decision Trees, Support Vector Machine, Artificial Neural Networks and Convolutional Neural Networks (CNN);
7. Feature extraction methods and feature selection methods;
8. Validation methods and classifier evaluation metrics;
9. Application of the methods discussed in different areas (economics, engineering, medicine, etc);
Evaluation Methodology
Written exam: 30% weight in the final grade (minimum grade of 40%);
Mini-practical exams and homework assignments with individual assessment and group projects: 70% weight in the final grade (minimum grade of 40%).
The minimum final grade is 10 out of 20.
The mini-tests will be held on dates to be determined (arranged in advance with the students), and the assignments and projects will be carried out throughout the semester with deadlines that will be defined according to the progress of the subjects taught.
These criteria and evaluation methods apply to all assessment periods.
Bibliography
- Bishop, C. (2006). Pattern recognition and machine learning. USA: Springer
- Géron, . e , . (2019). Hands-on Machine Learning with Sciki-Learn, Keras & TensorFlow. USA: O'Reilly
Teaching Method
- Lectures explaining the conceptual and technical aspects of the methods, accompanied by examples of Python programming;
- Problem-solving classes oriented towards programming;
- Project implementation.
Software used in class
Python IDE.
