Modulcode:	infSemMaLea-01a
Englische Bezeichnung:	Master Seminar - Machine Learning
Modulverantwortliche(r):	Prof. Dr. Peer Kröger
Turnus:	unregelmäßig (SS21 SS23)
Präsenzzeiten:	2S
ECTS:	5
Workload:	30 Std. Mitarbeit im Seminar, 120 Std. Selbststudium
Dauer:	ein Semester
Modulkategorien:	MSc-Inf-Sem (MSc Inf (21)) Sem (MSc Inf (15))
Lehrsprache:	Englisch
Voraussetzungen:

Kurzfassung:

The research in the area of Machine Learning (ML) has enormously grown in the last decades and subfields like Explainable AI (XAI), Representation Learning and ML-based NLP find next to established areas like Anomaly Detection high attention at top conferences. In this seminar, recent publications in these ML areas will be discussed.

Lernziele:

Studens will review papers concerning key concepts of different state-of-the-art (SOTA) methods from several ML topics and compare different approaches. At the end of the course you will present your findings and report them in a short paper.

Lehrinhalte:

Besides the exploration of SOTA ML approaches this seminar also gives a short introduction to application domains which can be optimized by these ML approaches.

Your papers will all be taken from the following list of topics that are relevant to our current research:

• Recent Developments in (Group) Anomaly Detection
• Explainable AI (XAI) in Anomaly Detection
• Representation Learning for Text and Graph Mining
• Artificial Intelligence for Sustainability and Safety
• Natural Language Processing (NLP) and Machine Learning for Opinion Mining

Typical publication process steps will be conducted during the seminar:

• structuring the paper content
• writing a paper according to scientific writing guidelines
• submitting the own paper and reviewing paper of others
• integration of review feedback and finalizing to a camera-ready paper version
• presenting the own seminar work at midterm and at the end of the seminar (final presenation)

Weitere Voraussetzungen:

Knowledge of scientific writing. Knowledge in machine learning (basics) and artificial neural networks.

Prüfungsleistung:

The final grade includes the evaluation of:

• the written paper
• the written paper reviews
• the paper presentation

Lehr- und Lernmethoden:

Verwendbarkeit:

Literatur:

The students will be given specific literature to work on in the seminar (changes from term to term).

Verweise:

Kommentar: