Dictionary-Based Low-Rank Approximations and the Mixed Sparse Coding Problem

Autor: Jeremy E. Cohen

Publikováno v: Frontiers in Applied Mathematics and Statistics, Vol 8 (2022)

Constrained tensor and matrix factorization models allow to extract interpretable patterns from multiway data. Therefore crafting efficient algorithms for constrained low-rank approximations is nowadays an important research topic. This work deals wi

Externí odkaz: https://doaj.org/article/e4767cce804443a185e0e6d369c0820c

A Flexible Optimization Framework for Regularized Matrix-Tensor Factorizations With Linear Couplings

Autor: Evrim Acar, Jeremy E. Cohen, Carla Schenker

Publikováno v: IEEE Journal of Selected Topics in Signal Processing
IEEE Journal of Selected Topics in Signal Processing, 2020, 15 (3), pp.506-521. ⟨10.1109/JSTSP.2020.3045848⟩
IEEE Journal of Selected Topics in Signal Processing, IEEE, 2020, 15 (3), pp.506-521. ⟨10.1109/JSTSP.2020.3045848⟩

International audience; Coupled matrix and tensor factorizations (CMTF) are frequently used to jointly analyze data from multiple sources, also called data fusion. However, different characteristics of datasets stemming from multiple sources pose man

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::7169cfc28546146b601538ea32c4d2e6
https://doi.org/10.1109/jstsp.2020.3045848

Dictionary-based Low-Rank Approximations and the Mixed Sparse Coding problem

Autor: Jeremy E. Cohen

Publikováno v: Frontiers in Applied Mathematics and Statistics
Frontiers in Applied Mathematics and Statistics, Frontiers Media S.A, 2022, 8, ⟨10.3389/fams.2022.801650⟩

International audience; Constrained tensor and matrix factorization models allow to extract interpretable patterns from multiway data. Therefore crafting efficient algorithms for constrained low-rank approximations is nowadays an important research t

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::918ad406b2a0ce3db2003ebe8f4c5379
http://arxiv.org/abs/2111.12399

Accelerating Block Coordinate Descent for Nonnegative Tensor Factorization

Autor: Jeremy E. Cohen, Le Thi Khanh Hien, Andersen Man Shun Ang, Nicolas Gillis

Publikováno v: Numerical Linear Algebra with Applications
Numerical Linear Algebra with Applications, 2021, 28 (5), ⟨10.1002/nla.2373⟩
Numerical Linear Algebra with Applications, Wiley, 2021, 28 (5), ⟨10.1002/nla.2373⟩

This paper is concerned with improving the empirical convergence speed of block-coordinate descent algorithms for approximate nonnegative tensor factorization (NTF). We propose an extrapolation strategy in-between block updates, referred to as heuris

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::38c51f4f63bf211908236411572add1a
https://hal.science/hal-03038513

Sparse Separable Nonnegative Matrix Factorization

Autor: Jeremy E. Cohen, Nicolas Gillis, Nicolas Nadisic, Arnaud Vandaele

Publikováno v: ECML PKDD 2020-European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases
ECML PKDD 2020-European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, Sep 2020, Ghent, Belgium. pp.1-20
Lecture Notes in Computer Science
Lecture Notes in Computer Science-Machine Learning and Knowledge Discovery in Databases
Machine Learning and Knowledge Discovery in Databases ISBN: 9783030676575
ECML/PKDD (1)
Machine Learning and Knowledge Discovery in Databases-European Conference, ECML PKDD 2020, Ghent, Belgium, September 14–18, 2020, Proceedings, Part I

We propose a new variant of nonnegative matrix factorization (NMF), combining separability and sparsity assumptions. Separability requires that the columns of the first NMF factor are equal to columns of the input matrix, while sparsity requires that

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::4958e3af71c844c44468374fa315c89e
https://hal.science/hal-02869490