Hierarchical alternating nonlinear least squares for nonnegative matrix factorization using rational functions.
Cécile HautecoeurFrançois GlineurLieven De LathauwerPublished in: EUSIPCO (2021)
Keyphrases
- nonnegative matrix factorization
- nonlinear least squares
- least squares
- levenberg marquardt
- negative matrix factorization
- trust region
- matrix factorization
- data representation
- newton method
- spectral clustering
- optical flow
- original data
- document clustering
- nearest neighbor
- image registration
- artificial neural networks
- machine learning
- principal component analysis
- bundle adjustment
- feature vectors
- objective function
- three dimensional