Position in the degree programme:

Compulsory elective

Workload:

5 LP / 4 SWS

Type of proof:

Portfolio

Learning outcomes / competences:

Students should understand basic questions and goals of machine learning, become familiar with special problem classes, such as supervised learning (classification and regression), develop important methods of machine learning and their scalable implementations and become familiar with concepts for evaluating learning methods. Students will independently apply what they have learnt to a complex task.

Contents:

Introduction and basic concepts, concept learning and version spaces, data preprocessing, case-based learning, decision trees, rule learning, support vector machines, extensions and meta-techniques, empirical evaluation of learning methods. Application in an own project, in design and mechanical engineering.