Text auf Deutsch
Course in German and English

Hours per Week:

4

Credit Points:

5

Prerequisites:

 

Frequency (WS/SS):

Sommersemester

Work Load:

 150h (60h class, 90h homework)

Study Programme Goals:

 understanding and implementing complex algorithms

Course Goals:

 Students know and understand fundamental theory and algorithms for time series analysis. They solve problems using scientific analysis, understand complex constraints and implement solutions using appropriate models.

Key Qualifications:

 structured problem solving, scientific analysis, implementation of complex algorithms

Course Contents:

 Students work on theory and algorithms for time series analysis: dynamic time warping, Markov chains, hidden Markov models, recurrent neural networks. Examples show pros and cons of each model. Prior knowledge in machine learning is recommended but not required.



Literature:

  • Niemann, H: Klassifikation von Mustern. 2. Überarbeitete Auflage, 2003
  • Goodfellow, I and Bengio,Y and Courville, A: Deep Learning, 2016 (available online: http://www.deeplearningbook.org/)
  • Huang, Acero, Hon: Spoken Language Processing: A Guide to Theory, Algorithm and System Development

Comments:

 The course materials are in English, the class will be given in English or German.

Assessment/Examination:

Paper (6 pages, IEEEtran style) and colloquium (15min), weighted 3:1

Admission Requirement:

 none

Auxiliary Means:

2

Lecturer(s):

Prof. Dr. Riedhammer





TH Nürnberg
Fakultät Informatik
Webmaster-IN



Root- Zertifikat

© 2020 Fakultät Informatik