|
|
Advanced Digital Signal Processing
PhD-course "Advanced Digital Signal Processing"
JUNE 10-14, 2002
Informatics and Mathematical Modelling
Build 321 Technical University of Denmark
This year the course will
focus on Independent Component Analysis
Aim:
To provide the participants with operational experience in
applications of independent component analysis (ICA) by
introducing the basic probabilistic concepts, applications
examples (e.g. from participants' own research) and a
selection of state-of-the-art algorithms.
Contents:
Background, motivation and examples of applications of ICA.
The probabilistic framework: likelihoods and the EM algorithm
Gaussian models: factor analysis, probabilistic PCA.
Temporal ICA: the Molgedey-Schuster algorithm
Maximum likelihood ICA: the Bell and Sejnowski algorithm
Bayesian ICA: Variational and advanced mean field algorithms
Applications: Images, text, speech, neuroimages.
Lectures: Lars Kai Hansen and Ole Winther
Matlab exercises: Thomas Kolenda, Lars Kai Hansen and Ole Winther
Type:
course at phd level, open university
Class schedule:
10-14. of June (morning lectures / afternoon exercises)
Scope and form:
Lectures, exercises (Matlab), short presentations by
participants.
Evaluation:
Project report
Examination:
Pass/fail
Registration
Ms Ulla Noerhave
Informatics and Mathematical Modelling
Technical University of Denmark
un@imm.dtu.dk
=====================================================================
Further information
Ole Winther
Associate Professor, Digital Signal Processing
Informatics and Mathematical Modelling (IMM)
Technical University of Denmark (DTU) http://www.imm.dtu.dk
Tel: +45 4525 3895 Homepage: http://eivind.imm.dtu.dk/staff/winther/
|