Joint-Diagonalizability-Constrained Multichannel Nonnegative Matrix Factorization Based on Multivariate Complex Sub-Gaussian Distribution.
Keigo KamoYuki KuboNorihiro TakamuneDaichi KitamuraHiroshi SaruwatariYu TakahashiKazunobu KondoPublished in: CoRR (2020)
Keyphrases
- gaussian distribution
- nonnegative matrix factorization
- probabilistic latent semantic indexing
- data representation
- matrix factorization
- maximum likelihood
- negative matrix factorization
- expectation maximization
- gaussian mixture model
- spectral clustering
- nonnegative matrix
- image segmentation
- co occurrence
- least squares
- knn
- training set
- feature space
- pattern recognition