Abstract

Non-negative sparse coding is a method for decomposing multivariate data into non-negative sparse components. In this paper we briefly describe the motivation behind this type of data representation and its relation to standard sparse coding and non-negative matrix factorization. We then give a simple yet efficient multiplicative algorithm for finding the optimal values of the hidden components. In addition, we show how the basis vectors can be learned from the observed data. Simulations demonstrate the effectiveness of the proposed method.

Citations

698	Learning the parts of objects by non-negative matrix factorization - Lee, Seung - 1999
683	Emergence of simple-cell receptive field properties by learning a sparse code for natural images - Olshausen, Field - 1996
464	Algorithms for non-negative matrix factorization - Lee, Seung - 2001
152	Positive matrix factorization: a nonnegative factor model with optimal utilization of error estimates of data values - Paatero, Tapper - 1994
48	Conditions for nonnegative independent component analysis - Plumbley - 2002
45	Development of low entropy coding in a recurrent network - Harpur, Prager - 1996
41	A multi-layer sparse coding network learns contour coding from natural images - Hoyer, Hyvärinen - 2002
27	Unmixing hyperspectral data - Parra, Spence, et al. - 2000
17	Low Entropy Coding with Unsupervised Neural Networks - Harpur - 1997
10	Modeling receptive fields with non-negative sparse coding - Hoyer