A Provably Correct and Robust Algorithm for Convolutive Nonnegative Matrix Factorization

Anthony Degleris; N. Gillis

DOI:10.1109/TSP.2020.2984163

Corpus ID: 189928495

A Provably Correct and Robust Algorithm for Convolutive Nonnegative Matrix Factorization

@article{Degleris2020APC,
  title={A Provably Correct and Robust Algorithm for Convolutive Nonnegative Matrix Factorization},
  author={Anthony Degleris and N. Gillis},
  journal={IEEE Transactions on Signal Processing},
  year={2020},
  volume={68},
  pages={2499-2512}
}

Anthony Degleris, N. Gillis
Published 2020
Computer Science, Mathematics
IEEE Transactions on Signal Processing

In this paper, we propose a provably correct algorithm for convolutive nonnegative matrix factorization (CNMF) under separability assumptions. CNMF is a convolutive variant of nonnegative matrix factorization (NMF), which functions as an NMF with additional sequential structure. This model is useful in a number of applications, such as audio source separation and neural sequence identification. While a number of heuristic algorithms have been proposed to solve CNMF, to the best of our knowledge… Expand

View on IEEE

Share This Paper

1 Citations

Background Citations

Figures and Topics from this paper

Diversified Bayesian Nonnegative Matrix Factorization

Maoying Qiao, Jun Yu, T. Liu, X. Wang, D. Tao
Computer Science
AAAI
2020

References

SHOWING 1-10 OF 41 REFERENCES

SORT BY

Semidefinite Programming Based Preconditioning for More Robust Near-Separable Nonnegative Matrix Factorization

N. Gillis, S. Vavasis
Mathematics, Computer Science
SIAM J. Optim.
2015

Majorization-minimization Algorithms for Convolutive NMF with the Beta-divergence

Dylan Fagot, H. Wendt, C. Févotte, P. Smaragdis
Computer Science
ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
2019