Course content

The course gives a basic introduction to data analysis and machine learning. It covers the learning regimes of supervised and unsupervised learning thoroughly, and a light introduction to reinforcement learning and explanation methods for machine learning models. The course work is project driven with focus on applications, using Python and commonly used machine learning libraries.

Learning outcome

Knowledge

The candidate has the knowledge of:

• fundamentals of machine learning with commonly used learning algorithms.

Skills

The candidate has:

• ability to analyse data sets, and train and evaluate machine learning models on data.
• can evaluate the adequateness of learning regimes based on the data.

General competencies

Candidate can:

• understand the basic principles of data analysis and machine learning.
• has the knowledge about the applicability and limitations of different contemporary learning algorithms.

Learning methods and activities

Teaching activities every week:

• Lectures using student-active learning methods such as teacher-led lectures, problem-based/case-based learning and solving practical problems in machine learning.
• Guided lab sessions will be conducted with teaching assistants with individual mentoring and assignment solving.

Mandatory assignments: Mandatory assignments will be provided and 90% of the assignments must be approved to qualify for the final exam.

Specific conditions

Recommended previous knowledge

• Basic programming knowledge in Python.
• Statistics from ISTG1003 or similar courses (ST1101/TMA4240/TMA4245)
• Linear algebra and Calculus from IMAG1002, IMAG2024 or similar courses (TMA4110/TMA4115/MA1201, TMA4105, MA1103)

Course materials

Hands-on Machine Learning with Scikit Learn, Keras and Tensorflow, 2022, Aurelien Geron

Credit reductions