Stephan Hasler, Heiko Wersing, Edgar Körner, "Combining Reconstruction and Discrimination with Class-Specific Sparse Coding", Neural Computation, vol. 19, no. 7, pp. 1897–1918, 2007.

Sparse coding is an important approach for the unsupervised learning of sensory features. In this contribution we present two new methods which extend the traditional sparse coding approach with supervised components. Our goal is to increase the suitability of the learned features for classification tasks while keeping most of their general representation capability. We analyze the effect of the new methods using a visualization on artificial data and discuss the results on two object test sets with regard to the properties of the found feature representation.