Abstract | ||
---|---|---|
Moreover, in order to have a physical interpretation, some constraints should be incorporated in the signal or image processing technique, such as the non-negativity of the solution. This paper deals with the non-negative pre-image problem in kernel machines, for nonlinear pattern recognition. While kernel machines operate in a feature space, associated to the used kernel function, a pre-image technique is often required to map back features into the input space. We derive a gradient-based algorithm to solve the pre-image problem, and to guarantee the non-negativity of the solution. Its convergence speed is significantly improved due to a weighted stepsize approach. The relevance of the proposed method is demonstrated with experiments on real datasets, where only a couple of iterations are necessary. |
Year | Venue | Field |
---|---|---|
2011 | European Signal Processing Conference | Radial basis function kernel,Pattern recognition,Kernel embedding of distributions,Tree kernel,Kernel principal component analysis,Polynomial kernel,Artificial intelligence,Kernel (image processing),Kernel method,Variable kernel density estimation,Mathematics,Machine learning |
DocType | ISSN | Citations |
Conference | 2076-1465 | 3 |
PageRank | References | Authors |
0.42 | 7 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Maya Kallas | 1 | 15 | 3.13 |
Paul Honeine | 2 | 367 | 34.41 |
Cédric Richard | 3 | 940 | 71.61 |
Clovis Francis | 4 | 34 | 11.20 |
Hassan Amoud | 5 | 36 | 8.61 |