Title :
Dictionary learning based dimensionality reduction for classification
Author :
Schnass, Karin ; Vandergheynst, Pierre
Author_Institution :
Signal Process. Inst., Swiss Fed. Inst. of Technol., Lausanne
Abstract :
In this article we present a signal model for classification based on a low dimensional dictionary embedded into the high dimensional signal space. We develop an alternate projection algorithm to find the embedding and the dictionary and finally test the classification performance of our scheme in comparison to Fisher´s LDA.
Keywords :
learning (artificial intelligence); matrix algebra; pattern classification; principal component analysis; classification; dictionary learning; dimensionality reduction; high dimensional signal space; low dimensional dictionary; Costs; Dictionaries; Glass; Image storage; Mouth; Nose; Projection algorithms; Signal processing; Space technology; Testing; alternate projections; classification; dictionary learning; dimensionality reduction; kernel-based learning;
Conference_Titel :
Communications, Control and Signal Processing, 2008. ISCCSP 2008. 3rd International Symposium on
Conference_Location :
St Julians
Print_ISBN :
978-1-4244-1687-5
Electronic_ISBN :
978-1-4244-1688-2
DOI :
10.1109/ISCCSP.2008.4537328
