117531
20190316234134.0
CONF
Dictionary Identifiability from Few Training Samples
2008
2008
Conference Papers
This article treats the problem of learning a dictionary providing sparse representations for a given signal class, via $ell^1$ minimisation, or more precisely the problem of identifying a dictionary $dico$ from a set of training samples $Y$ knowing that $Y = dico X$ for some coefficient matrix $X$. It provides a characterisation of coefficient matrices $X$ that allow to recover any orthonormal basis (ONB) as a local minimum of an $ell^1$ minimisation problem. Based on this characterisation it is shown that certain types of sparse random coefficient matrices will ensure local identifiability of the ONB with high probability.
lts2
sparse representation
dictionary learning
basis learning
recovery condition
random coefficients
Gribonval, Remi
240457
Schnass, Karin
168927
EUSIPCO
Lausanne
August 2008
Proc. EUSIPCO'08
URL
140321
http://infoscience.epfl.ch/record/117531/files/eusipco08.pdf
n/a
252392
LTS2
U10380
oai:infoscience.tind.io:117531
conf
STI
GLOBAL_SET
EPFL-CONF-117531
EPFL
REVIEWED
PUBLISHED
CONF