A new algorithm for learning overcomplete dictionaries

Abstract : In this paper, we propose a new algorithm for learning overcomplete dictionaries. The proposed algorithm is actually a new approach for optimizing a recently proposed cost function for dictionary learning. This cost function is regularized with a term that encourages low similarity between different atoms. While the previous approach needs to run an iterative limited-memory BFGS (l-BFGS) algorithm at each iteration of another iterative algorithm, our approach uses a closedform formula. Experimental results on reconstruction of a true underlying dictionary and designing a sparsifying dictionary for a class of autoregressive signals show that our approach results in both better quality and lower computational load.
Complete list of metadatas

Cited literature [14 references]  Display  Hide  Download

Contributor : Christian Jutten <>
Submitted on : Friday, September 27, 2013 - 4:09:47 PM
Last modification on : Tuesday, July 9, 2019 - 1:21:26 AM
Long-term archiving on : Friday, April 7, 2017 - 4:10:17 AM


Publisher files allowed on an open archive


  • HAL Id : hal-00867093, version 1


Mostafa Sadeghi, Massoud Babaie-Zadeh, Christian Jutten. A new algorithm for learning overcomplete dictionaries. 21st European Signal Processing Conference (EUSIPCO-2013), Sep 2013, Marrakech, Morocco. pp.EUSIPCO 2013 1569746047. ⟨hal-00867093⟩



Record views


Files downloads