Soft-competitive learning of sparse codes and its application to image reconstruction

Abstract

We propose a new algorithm for the design of overcomplete dictionaries for sparse coding, Neural Gas for Dictionary Learning (NGDL), which uses a set of solutions for the sparse coefficients in each update step of the dictionary. In order to obtain such a set of solutions, we additionally propose the bag of pursuits method (BOP) for sparse approximation… (More)
DOI: 10.1016/j.neucom.2011.02.002

10 Figures and Tables

Topics

  • Presentations referencing similar topics