Informações sobre a disciplina

Lehrinhalte
The course covers the following topics:
[list]
[*]Motivation, Applications
[*]Fundamentals
[list]
[*]definition of graphs, classes of graphs, properties of graphs, signals defined over graphs
[*]Adjecency matrix, Graph Laplacian, Graph shift operator
[*]Covariance matrix, conditional dependence, precision matrix
[/list]
[*]Graph signal processing
[list]
[*]Consensus, Diffusion
[*]Graph spectral analysis, Graph Fourier Transform
[*]Total variational norm, Graph Frequencies
[*]Bandlimited graph signals, smoothness
[*]Graph filters, Graph sampling theorem
[*]Applications
[/list]
[*]Network topology inference
[list]
[*]Link prediction
[*]Association network inference
[*]Tomographic network topology inference
[*]Pearson product-moment correlation
[*]Causality, Partial correlation
[*]Conditional independence graph
[*]Gaussian Markov Random Fields
[*]Graphical LASSO, Graphical LASSO with Laplacian constraint
[*]Applications
[/list]
[*]Graph analysis
[list]
[*]Subgraph identification
[*]Cliques identification
[/list]
[*]Optimization over graphs
[list]
[*]Average consensus, diffusion, exact diffusion
[*]Gradient tracking, push-sum algorithm, etc.
[*]Applications
[/list]
[*]Graph neuronal (convolutional) network
[/list]

Literature
[list]
[*]Lecture notes and slides can be downloaded here:
[list]
[*]www.nts.tu-darmstadt.de
[*]moodle
[/list]
[*]Further reading:
[list]
[*]Petar M. Djuric, Cédric Richard, Cooperative and Graph Signal Processing, Academic Press, 2018, ISBN 9780128136775.
[/list]
[/list]

Voraussetzungen
Basic knowledge in linear algebra and matrix analysis.