Aprendizagem Automática (Fcul)

Objectives

This course unit offers in-depth and comprehensive coverage of machine learning topics in order to provide students with the basic components of the discipline, so that they can understand the requirements, limitations and expected results of problematic data situations and provide techniques and tools for using them. of state-of-the-art methods to solve complex situations of varied scope.

Syllabus

Machine learning fundamentals; Tree models: decision trees; Linear models: linear regression, logistic regression and regularized models (LASSO and Ridge); Distance-based models: classification and regression of k-nearest neighbors; Probabilistic models: Naive Bayes; Kernel methods and support vector machines; Ensemble Methods. Validation and evaluation of models; Variable selection methods. Neural networks: the perceptron and backpropagation; An introduction to clustering with k-means and hierarchical clustering.

Prerequisites

Exposition classes, exercise resolution classes and/or laboratory classes.