Algorithms for Non-negative Matrix Factorization
D. D. Lee, H. Sebastian Seung · 2000
Non-negative matrix factorization (NMF) has previously been shown to be a useful decomposition for multivariate data. Two different multiplicative algorithms for NMF are analyzed. They differ only slightly in the multiplicative factor used in the update rules. One algorithm can be shown to minimize the conventional least squares error while the other minimizes the generalized Kullback-Leibler divergence. The monotonic convergence of both algorithms can be proven using an auxiliary function analogous to that used for proving convergence of the ExpectationMaximization algorithm. The algorithms can also be interpreted as diagonally rescaled gradient descent, where the rescaling factor is optimally chosen to ensure convergence. Introduction Unsupervised learning algorithms such as principal components analysis and vector quantization can be understood as factorizing a data matrix subject to different constraints. Depending upon the constraints utilized, the resulting factors can be shown ...
Cited by 5,378 works.
Authors
References
- Maximum Likelihood from Incomplete Data Via the EM Algorithm
- Eigenfaces for Recognition
- Numerical Recipes, The Art of Scientific Computing
- Principal Component Analysis
- Vector Quantization and Signal Compression
- Learning the parts of objects by non-negative matrix factorization
- Maximum Likelihood Reconstruction for Emission Tomography
- Bayesian-Based Iterative Method of Image Restoration*
- What Is the Goal of Sensory Coding?
- An iterative technique for the rectification of observed distributions