Name
Abstract | ||
|---|---|---|
We propose a preliminary investigation on the benefits and limitations of classifiers based on sparse representations. We specifically focus on the union of subspaces data model and examine binary classifiers built on a sparse non linear mapping (in a redundant dictionary) followed by a linear classifier. We study two common sparse non linear mappings (namely l0 and l1) and show that, in both cases, there exists a finite dictionary such that the classifier discriminates the two classes correctly. This result paves the way towards a better understanding of the increasingly popular classifiers based on sparse representations, and provides initial insights on appropriate dictionary design. |
Year | DOI | Venue |
|---|---|---|
2013 | 10.1109/ACSSC.2013.6810518 | Pacific Grove, CA |
Keywords | Field | DocType |
compressed sensing,signal classification,signal representation,binary classifiers,dictionary design,finite dictionary,linear classifier,redundant dictionary,sparse nonlinear mapping,sparse representations,subspaces data model,unions | K-SVD,Pattern recognition,Computer science,Sparse approximation,Feature extraction,Linear subspace,Artificial intelligence,Linear classifier,Classifier (linguistics),Data model,Machine learning,Binary number | Conference |
ISSN | ISBN | Citations |
1058-6393 | 978-1-4799-2388-5 | 1 |
PageRank | References | Authors |
0.38 | 11 | 2 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Alhussein Fawzi | 1 | 766 | 36.80 |
| Pascal Frossard | 2 | 3015 | 230.41 |