Generalized Correntropy Matching Pursuit: A novel, robust algorithm for sparse decomposition

Carlos A. Loza, Jose C. Principe

Producción científica: Capítulo del libro/informe/acta de congresoContribución a la conferenciarevisión exhaustiva

5 Citas (Scopus)

Resumen

We introduce a novel variation on the well-known Matching Pursuit (MP) algorithm. In particular, the sparse approximation problem is solved in a greedy scheme using estimated higher-order statistics as similarity measures instead of the somehow limited second-order statistics that perform optimally only under Gaussian assumptions. This is conveyed via the generalized correntropy (GC) function instead of the cross-correlation approach usually utilized in stochastic random processes applications. Additionally, extra flexibility is achieved by the GC parameters that control the behavior of the induced metric. The result is the robust Generalized Correntropy Matching Pursuit (GCMP) algorithm. Furthermore, we present results on two different frameworks dealing with detection and sparse approximation and highlight the robustness of this method in the presence of high-tailed impulsive noise.

Idioma originalInglés
Título de la publicación alojada2016 International Joint Conference on Neural Networks, IJCNN 2016
EditorialInstitute of Electrical and Electronics Engineers Inc.
Páginas1723-1727
Número de páginas5
ISBN (versión digital)9781509006199
DOI
EstadoPublicada - 31 oct. 2016
Publicado de forma externa
Evento2016 International Joint Conference on Neural Networks, IJCNN 2016 - Vancouver, Canadá
Duración: 24 jul. 201629 jul. 2016

Serie de la publicación

NombreProceedings of the International Joint Conference on Neural Networks
Volumen2016-October

Conferencia

Conferencia2016 International Joint Conference on Neural Networks, IJCNN 2016
País/TerritorioCanadá
CiudadVancouver
Período24/07/1629/07/16

Huella

Profundice en los temas de investigación de 'Generalized Correntropy Matching Pursuit: A novel, robust algorithm for sparse decomposition'. En conjunto forman una huella única.

Citar esto