Information on individual educational components (ECTS-Course descriptions) per semester

  
Degree programme:Master Computer Science
Type of degree:FH Master´s Degree Programme
 Full-time
 Winter Semester 2024
  

Course unit titleNeural Networks
Course unit code024913110503
Language of instructionGerman
Type of course unit (compulsory, optional)Elective
Semester when the course unit is deliveredWinter Semester 2024
Teaching hours per week2
Year of study2024
Level of course unit (e.g. first, second or third cycle)Second Cycle (Master)
Number of ECTS credits allocated4
Name of lecturer(s)Sebastian HEGENBART


Prerequisites and co-requisites

Linear algebra, analysis, probability theory and statistics

Course content
  • Artificial neuron models
  • Perceptron
  • Adaline
  • Single-layer neural networks (NN) and corresponding (monitored) learning techniques
  • Madaline
  • Feed-forward neural network
  • Error back propagation
  • non-linear optimization for NN learning
  • generalization: coping with the bias-variance dilemma
Learning outcomes

The students can

  • describe basic methods and algorithms from the field of supervised learning with neural networks. This also includes an understanding of the corresponding techniques of nonlinear optimization.
  • apply these algorithms and methods and analyze the results.
Planned learning activities and teaching methods

Lectures with integrated exercises.

Assessment methods and criteria
  • 75% exam (must be positive)
  • 25% exercise (must be positive)

For a positive grade, a minimum of 50% of the possible points must be achieved in each part of the examination.




    
Comment

    
 None

    
Recommended or required reading

    
Use of Octave or MatLab.


    

  •  Engelbrecht, Andries P. (2007): Computational Intelligence: An Introduction. 2nd edition Chichester, England; Hoboken, NJ: John Wiley & Sons.
    • 

  • Bishop, Christopher M. (1995): Neural Networks for Pattern Recognition. Oxford University Press.
    • 

  • Goodfellow, Ian; Bengio, Yoshua; Courville, Aaron (2017): Deep Learning. The MIT Press
    • 



 

    
Mode of delivery (face-to-face, distance learning)

    
Face-to-face event with recording of the lecture








Winter Semester 2024go Top