An Alternating Rank-K Nonnegative Least Squares Framework (ARkNLS) for Nonnegative Matrix Factorization.
Delin ChuWenya ShiSrinivas EswarHaesun ParkPublished in: CoRR (2020)
Keyphrases
- nonnegative matrix factorization
- least squares
- probabilistic latent semantic indexing
- data representation
- matrix decomposition
- sparsity constraints
- search engine
- data matrix
- matrix factorization
- spectral clustering
- principal component analysis
- negative matrix factorization
- optical flow
- pairwise
- pattern recognition
- objective function
- data mining