Positive tensor factorization
Abstract
A novel fixed point algorithm for positive tensor factorization (PTF) is introduced. The update rules efficiently minimize the reconstruction error of a positive tensor over positive factors. Tensors of arbitrary order can be factorized, which extends earlier results in the literature. Experiments show that the factors of PTF are easier to interpret than those produced by methods based on the singular value decomposition, which might contain negative values. We also illustrate the tendency of PTF to generate sparsely distributed codes.
- Publication:
-
Pattern Recognition Letters
- Pub Date:
- 2001
- DOI:
- 10.1016/S0167-8655(01)00070-8
- Bibcode:
- 2001PaReL..22.1255W
- Keywords:
-
- PCA;
- SVD;
- Positive matrix factorization;
- Feature extraction