FIDOS: A generalized Fisher based feature extraction method for domain shift
Impacto
Scholar |
Otros documentos de la autoría: Dinh, Cuong V.; Duin, Robert P. W.; Piqueras Salazar, Ignacio Jaime; Loog, Marco
Metadatos
Mostrar el registro completo del ítemcomunitat-uji-handle:10234/9
comunitat-uji-handle2:10234/7038
comunitat-uji-handle3:10234/8634
comunitat-uji-handle4:
INVESTIGACIONEste recurso está restringido
http://dx.doi.org/10.1016/j.patcog.2013.02.011 |
Metadatos
Título
FIDOS: A generalized Fisher based feature extraction method for domain shiftFecha de publicación
2013Editor
ElsevierISSN
0031-3203Tipo de documento
info:eu-repo/semantics/articleVersión de la editorial
http://ac.els-cdn.com/S0031320313001039/1-s2.0-S0031320313001039-main.pdf?_tid=2 ...Palabras clave / Materias
Resumen
Traditional pattern recognition techniques often assume that the data sets used for training and testing follow the same distribution. However, this assumption is usually not true for many real world problems as data ... [+]
Traditional pattern recognition techniques often assume that the data sets used for training and testing follow the same distribution. However, this assumption is usually not true for many real world problems as data from the same classes but different domains, e.g., data are collected under different conditions, may show different characteristics. We introduce FIDOS, a generalized FIsher based method for DOmain Shift problem, that aims at learning invariant features across domains in a supervised manner.
Different from classical Fisher feature extraction, FIDOS aims to minimize not only the within-class scatter but also the difference in distributions between domains. Therefore, the subspace constructed by FIDOS reduces the drift in distributions among different domains and at the same time preserves the discriminants across classes. Another advantage of FIDOS over classical Fisher is that FIDOS extracts more features when multiple source domains are available in the training set; this is essential for a good classification especially when the number of classes is small. Experimental results on both artificial and real data and comparisons with other methods demonstrate the efficiency of our method in classifying objects under domain shift situations. [-]
Publicado en
Pattern Recognition, 2013, Septiembre, Vol. 46, nº 9Derechos de acceso
http://rightsstatements.org/vocab/CNE/1.0/
info:eu-repo/semantics/restrictedAccess
info:eu-repo/semantics/restrictedAccess
Aparece en las colecciones
- LSI_Articles [361]