Convergence-guaranteed multiplicative algorithms for nonnegative matrix factorization with β-divergence

@article{Nakano2010ConvergenceguaranteedMA,
  title={Convergence-guaranteed multiplicative algorithms for nonnegative matrix factorization with β-divergence},
  author={Masahiro Nakano and Hirokazu Kameoka and Jonathan Le Roux and Yu. Kitano and Nobutaka Ono and Shigeki Sagayama},
  journal={2010 IEEE International Workshop on Machine Learning for Signal Processing},
  year={2010},
  pages={283-288}
}
This paper presents a new multiplicative algorithm for nonnegative matrix factorization with β-divergence. The derived update rules have a similar form to those of the conventional multiplicative algorithm, only differing through the presence of an exponent term depending on β. The convergence is theoretically proven for any real-valued β based on the auxiliary function method. The convergence speed is experimentally investigated in comparison with previous works. 
Highly Cited
This paper has 142 citations. REVIEW CITATIONS
64 Citations
17 References
Similar Papers

Citations

Publications citing this paper.
Showing 1-10 of 64 extracted citations

143 Citations

0102030'11'13'15'17
Citations per Year
Semantic Scholar estimates that this publication has 143 citations based on the available data.

See our FAQ for additional information.

References

Publications referenced by this paper.
Showing 1-10 of 17 references

On the use of the beta divergence for musical source separation

  • D. FitzGerald, M. Cranitch, E. Coyle
  • Proc. of Irish Signals and Systems Conference…
  • 2009
2 Excerpts

Similar Papers

Loading similar papers…