We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

stat.ML

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Statistics > Machine Learning

Title: Dictionary-based Tensor Canonical Polyadic Decomposition

Abstract: To ensure interpretability of extracted sources in tensor decomposition, we introduce in this paper a dictionary-based tensor canonical polyadic decomposition which enforces one factor to belong exactly to a known dictionary. A new formulation of sparse coding is proposed which enables high dimensional tensors dictionary-based canonical polyadic decomposition. The benefits of using a dictionary in tensor decomposition models are explored both in terms of parameter identifiability and estimation accuracy. Performances of the proposed algorithms are evaluated on the decomposition of simulated data and the unmixing of hyperspectral images.
Subjects: Machine Learning (stat.ML)
Journal reference: IEEE Trans. on Signal Processing 66 (7), pp. 1876-1889, 2018
DOI: 10.1109/TSP.2017.2777393
Cite as: arXiv:1704.00541 [stat.ML]
  (or arXiv:1704.00541v2 [stat.ML] for this version)

Submission history

From: Jérémy Emile Cohen [view email]
[v1] Mon, 3 Apr 2017 12:03:39 GMT (3368kb,D)
[v2] Wed, 8 Nov 2017 10:10:47 GMT (3381kb,D)

Link back to: arXiv, form interface, contact.