Related Experiment Videos

Learning overcomplete representations.

M S Lewicki1, T J Sejnowski

  • 1Computer Science Dept. and Center for the Neural Basis of Cognition, Carnegie Mellon Univ., 115 Mellon Inst., Pittsburgh, PA 15213, USA.

Neural Computation
|January 15, 2000
PubMed
Summary

This study introduces a new algorithm for learning overcomplete bases, enhancing data representation efficiency and robustness. The method improves signal approximation and coding efficiency, generalizing independent component analysis for noise-robust reconstruction and blind source separation.

Related Concept Videos