|
Šifra:
|
11512
|
|
ECTS:
|
6
|
Lecturers in charge:
|
Prof. dr. sc.
Damir Seršić
Doc. dr. sc.
Ivica Kopriva
|
English level:
1,1,1
|
All teaching activities in the course will be held on English. This level includes courses with multiple groups (i.e., all teaching will be held strictly in Croatian for Croatian groups, and strictly in English for English groups).
|
Description:
|
Mathematical preliminaries: stochastic processes, gradients and optimization methods, information theory. Blind separation of signals by principal (PCA) and independen (ICA) component analysis:conditions for uniqueness of the solution. Information-theoretic approach to ICA. Equivalence between ICA methods based on minimum of mutual information and maximum negentropy. Single and multichannel blind deconvolution in timr and frequency domains. Underdetermined blind source separation: sparse component analysis and nonnegative matrix factorizations.
|
Literature:
|
- A. Hyvärinen, J. Karhunen, E. Oja: Independent componenet analysis, John Wiley, 2002.
- A. Cichocki, S. Amari, Adaptive Blind Suignal and Image processing, John Wiley, 2002.
- T. - M. Huang, V. Kecman, I. Kopriva: kernel based Algorithms for Mining Huge Dana Sets: Supervised, Semi-supervised and Unsupervised Learning, Springer Series: Studies in Computational Intelligence, Vol. 17, XVI, ISBN: 3-540-31681-7, 2006.
|
|
