Dictionary Learning and Sparse Coding for Third-order Super-symmetric Tensors

Abstract : Super-symmetric tensors – a higher-order extension of scatter matrices – are becoming increasingly popular in machine learning and computer vision for modelling data statistics, co-occurrences, or even as visual descriptors. However, the size of these tensors are exponential in the data dimensionality, which is a significant concern. In this paper, we study third-order super-symmetric tensor descriptors in the context of dictionary learning and sparse coding. Our goal is to approximate these tensors as sparse conic combinations of atoms from a learned dictionary, where each atom is a symmetric positive semi-definite matrix. Apart from the significant benefits to tensor compression that this framework provides, our experiments demonstrate that the sparse coefficients produced by the scheme lead to better aggregation of high-dimensional data, and showcases superior performance on two common computer vision tasks compared to the state-of-the-art.
Type de document :
Autre publication
NIPS submission. 2015, pp.13
Liste complète des métadonnées

Littérature citée [32 références]  Voir  Masquer  Télécharger

Contributeur : Piotr Koniusz <>
Soumis le : vendredi 11 septembre 2015 - 16:12:02
Dernière modification le : mercredi 26 avril 2017 - 14:03:04
Document(s) archivé(s) le : mardi 29 décembre 2015 - 00:43:01


Fichiers produits par l'(les) auteur(s)




Piotr Koniusz, Anoop Cherian. Dictionary Learning and Sparse Coding for Third-order Super-symmetric Tensors. NIPS submission. 2015, pp.13. 〈hal-01197410〉



Consultations de la notice


Téléchargements de fichiers