Ungrounded Independent Non-Negative Factor Analysis

    •  Raj, B.; Wilson, K.W.; Krueger, A.; Haeb-Umbach, R., "Ungrounded Independent Non-Negative Factor Analysis", Annual Conference of the International Speech Communication Association, September 2010, pp. 330-333.
      BibTeX Download PDF
      • @inproceedings{Raj2010sep,
      • author = {Raj, B. and Wilson, K.W. and Krueger, A. and Haeb-Umbach, R.},
      • title = {Ungrounded Independent Non-Negative Factor Analysis},
      • booktitle = {Annual Conference of the International Speech Communication Association},
      • year = 2010,
      • pages = {330--333},
      • month = sep,
      • url = {}
      • }
  • Research Area:

    Speech & Audio

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.