Seminar for PhD Students
- TISS: (link)
- contact: David Penz
Learning outcomes
After successful completion of the course, students are able to describe basic concepts of machine learning (incl. data preparation, selection of suitable algorithms, evaluation) and apply them to real-world problems.
Professional and methodological competences: After positive completion of the module, students are able to
- develop a suitable strategy for dealing with a given problem (selection of algorithms and methods),
- work out and apply the basics and formal concepts of machine learning,
- develop a suitable strategy for processing real data,
- define an evaluation concept.
Cognitive and practical competences: After positive completion of the module, students are able to
- understand existing problems and their underlying concepts,
- analyse data sets and prepare them for correct use,
- apply different algorithms and solution approaches to real data,
- correctly evaluate applied methods and interpret results.
Social competences and personal competences: After positive completion of the module, students are able to independently analyse problems, apply and evaluate appropriate methods and interpret results.
Subject of course
Planned contents are:
- Introduction, history and taxonomy
- Basic concepts of machine learning (error bounds, data preparation and evaluation methods) and applications
- Rule-based classification and regression
- Clustering and dimensionality reduction
- Learning theory
- Kernel methods
- Probabilistic models
- Ensemble Methods
- Deep Learning
- Online, Active and Reinforcement Learning
- Outlook including fairness and ethics in machine learning
Teaching methods
A mix of introductory online lectures (recorded and/or live), exercises with formative feedback and some live (online) sessions where the assigments are discussed.