Strategies to learn computationally efficient and compact dictionaries

Luc Le Magoarou 1 Rémi Gribonval 1
1 PANAMA - Parcimonie et Nouveaux Algorithmes pour le Signal et la Modélisation Audio
IRISA-D5 - SIGNAUX ET IMAGES NUMÉRIQUES, ROBOTIQUE, Inria Rennes – Bretagne Atlantique
Abstract : Dictionary learning is a branch of signal processing and machine learning that aims at expressing some given training data matrix as the multiplication of two factors: one dense matrix called dictionary and one sparse matrix being the representation of the data in the dictionary. The sparser the representation, the better the dictionary. However, manipulating the dictionary as a dense matrix can be computationally costly both in the learning process and later in the usage of this dictionary, thus limiting dictionary learning to relatively small-scale problems. In this paper we consider a general structure of dictionary allowing faster manipulation, and give an algorithm to learn such dictionaries over training data, as well as preliminary results showing the interest of our approach.
Complete list of metadatas

Cited literature [12 references]  Display  Hide  Download

https://hal.inria.fr/hal-01010766
Contributor : Luc Le Magoarou <>
Submitted on : Friday, June 20, 2014 - 2:18:51 PM
Last modification on : Thursday, November 15, 2018 - 11:58:45 AM
Long-term archiving on : Saturday, September 20, 2014 - 10:56:40 AM

File

iTWIST2014_FINAL.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01010766, version 1

Citation

Luc Le Magoarou, Rémi Gribonval. Strategies to learn computationally efficient and compact dictionaries. International Traveling Workshop on Interactions between Sparse models and Technology (iTWIST), Aug 2014, Namur, Belgium. ⟨hal-01010766⟩

Share

Metrics

Record views

1309

Files downloads

152