Reference : Global Optimality Bounds for ICA Algorithms
Scientific congresses, symposiums and conference proceedings : Paper published in a book
Physical, chemical, mathematical & earth Sciences : Mathematics
http://hdl.handle.net/10993/28005
Global Optimality Bounds for ICA Algorithms
English
Colombo, Nicolo mailto [University of Luxembourg > Luxembourg Centre for Systems Biomedicine (LCSB) > >]
Thunberg, Johan mailto [University of Luxembourg > Luxembourg Centre for Systems Biomedicine (LCSB) > >]
Goncalves, Jorge mailto [University of Luxembourg > Luxembourg Centre for Systems Biomedicine (LCSB) > >]
2016
22nd International Symposium on Mathematical Theory of Networks and Systems
Yes
Minneapolis
USA
22nd International Symposium on Mathematical Theory of Networks and Systems
July 12-15
[en] Independent Component Analysis ; spectral relaxation ; simultaneous diagonalization
[en] Independent Component Analysis is a popular statistical method for separating a multivariate signal into additive components. It has been shown that the signal separation problem can be reduced to the joint diagonalization of the matrix slices of some higher-order cumulants of the signal. In this approach, the unknown mixing matrix can be computed directly from the obtained joint diagonalizer. Various iterative algorithms for solving the non-convex joint diagonalization problem exist, but they usually lack global optimality guarantees. In this paper, we introduce a procedure for computing an optimality gap for local optimal solutions. The optimality gap is then used to obtain an empirical error bound for the estimated mixing matrix. Finally, a class of simultaneous matrix decomposition problems that admit such relaxation procedure is identified.
Fonds National de la Recherche - FnR
Researchers
http://hdl.handle.net/10993/28005
FnR ; FNR8864515 > Johan Thunberg > > Set Convergence in Nonlinear Multi-Agent Systems > 01/02/2015 > 31/01/2017 > 2014

File(s) associated to this reference

Fulltext file(s):

FileCommentaryVersionSizeAccess
Open access
MTNS16_0138_FI.pdfPublisher postprint591.27 kBView/Open

Bookmark and Share SFX Query

All documents in ORBilu are protected by a user license.