Advanced Machine Learning

Recommended Prerequisites

Linear Algebra, Calculus, Probabilities and Statistics, Introduction to Machine Learning, Programming.

Teaching Methods

During the lectures (T) the concepts, the theories, the algorithms will be presented and discussed. In the lan classes (PL) students will consolidate what was learned in T. The practical work will be done under the supervision of the teacher. Grading will be based on three components: (1) study and a written synthesis of a research paper; (2) a small project involving the techniques and/or a practical problem; (3) a written exam to assess students' knowledge about the subject of ML.

Learning Outcomes

This course will present and discuss with rigor the most recent algorithmic advances in machine learning, including ensembles, deep learning and evolutionary machine learning. Moreover, some other important aspects like feature engineering and the design of experiments and data analysis will be discussed.

By the end of the course, the student will have a general overview and a practical experience on how to transform data into knowledge, and will master the most recent techniques of machine learning. He/she will be capable of design, implement, test and validate solutions for real world problems that require machine learning.

Last but not the least, the student will consolidate his/her communicational competences of analysis and synthesis, written and spoken, and of  group work.

Work Placement(s)

No

Syllabus

1. Introduction: the data science problem
2. Fundamentals Algorithms
3. Ensembles: bagging, random forests, boosting, stacking
4. Features: transformations, construction, selection
5. Reinforcement Learning
6. Deep Learning
7. Evolutionary Machine Learning
8. Design of Experiments and Data Analysis
9. Applications

Head Lecturer(s)

Ernesto Jorge Fernandes Costa

Assessment Methods

Assessment
Research work: 10.0%
Other: 10.0%
Project: 30.0%
Exam: 50.0%

Bibliography

- Peter Flach, Machine Learning: the art and science of algorithms that make sense of data, Cambridge University Press,2012.
- Ethem Alpaydin, Introduction to Machine Learning (2nd Edition), MIT Press. 2010.
- Richard S. Sutton and Andrew G. Barto, Reinforcement  Learning: an introduction (2nd Edition), MIT Press, 2018.
- Ian Goodfellow, Yoshua Bengio and Aaron Courville, Deep Learning, MIT Press 2016.
- Hitoshi Iba, Evolutionary Approaches to Machine Learning and Deep Neural Networks, Springer, 2018.