Aprendizagem Automática em Bioengenharia

Objectives

This course aims to provide insight and knowledge on state-of-the-art machine learning and data mining techniques, and their broad application to a diversity of real-world datasets and problems. Application examples addressed in the course include sensor-based, web-based, computer vision, biomechanics, biological systems, bioinformatics, human-centered health monitoring, prediction, and disease prevention. Students completing the course are expected to: 1) understand the fundamental concepts, and challenges of machine learning and data mining techniques; 2) have a clear understanding of its applicability and empowerment over a broad range of areas transversal to most engineering courses; 3) be able to solve real-world problems in the several scientific areas and application domains, with a proper understanding of what the tools mean, when and how to apply them, and critically evaluate and compare the solutions provided

Syllabus

1.Taxonomies of learning techniques. Machine learning and datamining. Neural Networks and Deep Learning. 2.Types of data. Review of probability and information theory concepts. (Dis)similarity measures and (dis)similarity- based data representation. Learning data representation. 3.Supervised learning techniques. Geometric, probabilistic, and hybrid approaches. Deep Learning. Classifier fusion techniques. 4.Evaluation of classifier performance. Cross- validation and bootrapping techniques. 5.The curse of dimensionality and reduction of feature space. Feature selection and space transformation techniques. 6.Big data and data labeling. Semi-supervised-learning. Active learning and Interactive Machine learning. Transfer Learning. 7.Unsupervised learning and clustering. Clustering Ensembles and information fusion.

Prerequisites

Without prerequisites to attend the curricular unit.

Cross Competence Component

Writing and oral presentation skills.

Laboratorial Component

The first 4 laboratory sessions aim at apllying some of the algorithms and techniques addressed in the theoretical lectures. The remaining lab sessions are used for giving support and supervision of the work associated with the groups' projects.

Programming And Computing Component

The curricular unit explores and applies a substantial amount of computation and programming in the context of team projects, addressing real-world problems, developed along the semester, as well as the 4 lab sessions addressing machine learning techniques and algorithms. Overall, these contribute to 40% of the evaluation.

Ethical Principles

All members of a group are responsible for the group’s work In any assessment every student shall honestly disclose any help received and sources used. In an oral assessment, every student shall be alble to present and answer questions about the entire assignment and solution.