11th Annual Conference of the International Speech Communication Association

Makuhari, Chiba, Japan
September 26-30. 2010

Ungrounded Independent Non-Negative Factor Analysis

Bhiksha Raj (1), Kevin W. Wilson (2), Alexander Krueger (3), Reinhold Haeb-Umbach (3)

(1) Carnegie Mellon University, USA
(3) Universität Paderborn, Germany

We describe an algorithm that performs regularized non-negative matrix factorization (NMF) to find independent components in non- negative data. Previous techniques proposed for this purpose require the data to be grounded, with support that goes down to 0 along each dimension. In our work, this requirement is eliminated. Based on it, we present a technique to find a low-dimensional decomposition of spectrograms by casting it as a problem of discovering independent non-negative components from it. The algorithm itself is implemented as regularized non-negative matrix factorization (NMF). Unlike other ICA algorithms, this algorithm computes the mixing matrix rather than an unmixing matrix. This algorithm provides a better decomposition than standard NMF when the underlying sources are independent. It makes better use of additional observation streams than previous nonnegative ICA algorithms.

Full Paper

Bibliographic reference.  Raj, Bhiksha / Wilson, Kevin W. / Krueger, Alexander / Haeb-Umbach, Reinhold (2010): "Ungrounded independent non-negative factor analysis", In INTERSPEECH-2010, 330-333.