A multiplicative algorithm for convolutive non-negative matrix factorization based on squared euclidean distance

Wenwu Wang, Andrzej Cichocki, Jonathaon A. Chambers

Research output: Contribution to journalArticlepeer-review

52 Citations (SciVal)

Abstract

Using the convolutive nonnegative matrix factorization (NMF) model due to Smaragdis, we develop a novel algorithm for matrix decomposition based on the squared Euclidean distance criterion. The algorithm features new formally derived learning rules and an efficient update for the reconstructed nonnegative matrix. Performance comparisons in terms of computational load and audio onset detection accuracy indicate the advantage of the Euclidean distance criterion over the Kullback-Leibler divergence criterion.

Original languageEnglish
Pages (from-to)2858-2864
Number of pages7
JournalIEEE Transactions on Signal Processing
Volume57
Issue number7
DOIs
Publication statusPublished - 2009
Externally publishedYes

Keywords

  • Audio object separation
  • Convolutive nonnegative matrix factorization
  • Multiplicative algorithm
  • Squared euclidean distance

Fingerprint

Dive into the research topics of 'A multiplicative algorithm for convolutive non-negative matrix factorization based on squared euclidean distance'. Together they form a unique fingerprint.

Cite this