Modulinformationssystem Informatik

 

Master Seminar - Deep Learning URL PDF XML

Modulcode: infSemDeLea-01a
Englische Bezeichnung: Master Seminar - Deep Learning
Modulverantwortliche(r): Prof. Dr.-Ing. Sven Tomforde
Turnus: unregelmäßig (WS21/22 WS22/23 WS23/24 SS24 WS24/25)
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: Info

Kurzfassung:

The KickOff meeting for the seminar is scheduled for Tuesday, October 17th at 2 pm in room 0309b (Hermann-Rodewald-Str. 3, i.e. in the floor of the INS group). Please join us there!

Deep learning is a method of machine learning (and thus a subfield of artificial intelligence) that uses artificial neural networks (KNNs) with numerous hidden layers between the input layer and the output layer, thereby forming an extensive internal structure. In recent years, Deep Learning has established itself as a standard method for solving a wide variety of learning problems. In this seminar, current developments in the field of Deep Learning are discussed.

Lernziele:

Students learn to independently acquire complex subject matter using the compact presentation common in scientific publications (i.e. papers, articles or textbooks) and to prepare it in an understandable form.

Lehrinhalte:

The subject of the seminar are applications of Deep Learning and novel solutions for optimising the methods.

The focus in each semester is on different developments in the field of Deep Learning. Students will examine and present different basic systems in addition to basic technologies in the respective area. This also includes a classification in the superordinate framework of Deep Learning methods.

Conceptually, the course replicates a prototypical academic publication process: Creation of a concept, creation of a publication, submission and review using the conference management system, preparation of the camera-ready version and submission, presentation at the conference including moderation.

Weitere Voraussetzungen:

Basic knowledge of scientific work.

Prior knowledge in machine learning (basics) and artificial neural networks.

Prüfungsleistung:

The grade is composed of the following aspects:
  • Elaboration (i.e. paper according to a template)
  • Reviews
  • Organisation (role of the session chair, meeting deadlines)
  • 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: