Course description

The course introduces the main concepts and tools in machine learning which are necessary for professional work and research in data analytics. The course presents machine learning mainly from a probabilistic framework, but successful non-probabilistic methods are also covered.

The course is given in the following formats:

• Master level course 732A99: the students are supposed to participate in the sessions from block 1 and 2. Obligatory attendance of the seminars . Examination by submission of the lab reports and a computer-based exam
• Single subject course 732A68: the students are supposed to participate in the sessions from block 1 and 2. Obligatory attendance of the seminars. Examination by submission of the lab reports and a computer-based exam
• Engineering course TDDE01: the students are supposed to participate in the sessions  from block 1. Obligatory attendance of the seminars. Examination by submission of the lab reports and a computer-based exam

The course is divided into topics, where each topic includes lectures, a computer lab and a follow-up seminar.

Course literature

- Machine learning - a first course for engineers and scientists (MLFC) by Lindholm et al. Available for free here
- The Elements of Statistical Learning (ESL) by T. Hastie, R. Tibshirani and J. Friedman, Second Edition, ISBN 9780387848587, available for free here
- Slides
- Other material distributed during the course.

Topic 1: Basic concepts in machine learning. Software. Classification and regression.

Teacher: Oleg Sysoev.
Read: MLFC 1.1-2.2, ESL 2.5 (lecture la), MLFC 3.1-3.3 (lecture 1d)
Tutorial 1: here
Lecture 1a: Introduction to Machine Learning. Slides in LISAM
Lecture 1b: Basics of Statistics. Slides in LISAM
Lecture 1c: Introduction to R. Slides in LISAM
Lecture 1d: Slides in LISAM
 

Topic 2: Model selection and dimensionality reduction.

Teacher: Oleg Sysoev.
Read: MLFC 4.1-4.5 (lecture 2a), MLFC 2.3 (lecture 2b), MLFC 10.4 (lecture 2c), MLFC 5.1-5.6 (lecture 2d)
Tutorial 2: here
Lecture 2a: Slides in LISAM
Lecture 2b: Slides in LISAM
Lecture 2c: Slides in LISAM
Lecture 2d: Slides in LISAM

Topic 3: Kernel methods and support vector machines. Neural networks and deep learning.

Teacher: Jose M. Pena.
Read: MLFC 8.1-8.2 (lecture 3a), MLFC 8.3 and 8.5 (lecture 3b), MLFC 6.1-6.2 (lecture 3c), MLFC 6.3-6.4 (lecture 3d).
Lecture 3a: Slides in LISAM
Lecture 3b: Slides in LISAM
Lecture 3c: Slides in LISAM
Lecture 3d: Slides in LISAM

Topic 1: Additional topics in Machine Learning: ensemble methods, mixture models.

Teacher: Jose M. Pena.
Read: MLFC 7.1-7.4 (lecture a), MLFC 10.1-10.2 (lecture b)
Lecture a: Slides in LISAM
Lecture b: Slides in LISAM