Generalization Bounds for K-Dimensional Coding Schemes in Hilbert Spaces

We give a bound on the expected reconstruction error for a general coding method where data in a Hilbert space are represented by finite dimensional coding vectors. The result can be specialized to Kmeans clustering, nonnegative matrix factorization and the sparse coding techniques introduced by Olshausen and Field.

• Type: Conference
• Year: 2008
• Where: ALT
• Authors: Andreas Maurer, Massimiliano Pontil
• Keywords: ALT 2008, Dimensional Coding Vectors, General Coding Method, Machine Learning, Sparse Coding Techniques