Module PY4230-KP06

Duration

1 Semester

Turnus of offer

each summer semester

Credit points

6

Course of studies, specific fields and terms:

• Master Psychology - Cognitive Systems 2022, compulsory, psychology

Classes and lectures:

• CS4405-V: Neuroinformatik (lecture, 2 SWS)
• CS4230-S: Computational Neurosciences (seminar / exercises, 2 SWS)

Workload:

• 60 hours in-classroom work
• 90 hours private studies
• 30 hours exam preparation

Contents of teaching:

• The human brain and abstract neuron models
• Learning with a single neuron: * Perceptrons * Max-Margin Classification * LDA and logistic Regression
• Network architectures: * Hopfield-Networks * Multilayer-Perceptrons * Deep Learning
• Recurrent networks (LSTM, GPU)
• Unxupervised Learning: * k-means, Neural Gas and SOMs * PCA & ICA * Sparse Coding

Qualification-goals/Competencies:

• The students are able to understand the principle function of a single neuron and the brain as a whole.
• They know abstract neuronal models and they are able to name practical applications for the different variants.
• They are able to derive a learning rule from a given error function.
• They are able to apply (and implement) the proposed learning rules and approaches to solve unknown practical problems.

Grading through:

• written exam

Responsible for this module:

• Prof. Dr. rer. nat. Thomas Martinetz

Teacher:

• Institute for Neuro- and Bioinformatics
• Prof. Dr. rer. nat. Thomas Martinetz
• N.N.

Literature:

• S. Haykin : Neural Networks London: Prentice Hall, 1999
• J. Hertz, A. Krogh, R. Palmer : Introduction to the Theory of Neural Computation Addison Wesley, 1991
• T. Kohonen : Self-Organizing Maps Berlin: Springer, 1995
• H. Ritter, T. Martinetz, K. Schulten : Neuronale Netze: Eine Einführung in die Neuroinformatik selbstorganisierender Netzwerke Bonn: Addison Wesley, 1991
• Ian Goodfellow, Yoshua Bengio, Aaron Courville : Deep Learning MIT Press 2016

Language:

• offered only in German

Notes:

Last Updated:

04.12.2024