Joint-diagonalizability-constrained multichannel nonnegative matrix factorization based on time-variant multivariate complex sub-Gaussian distribution.
Keigo KamoYoshiki MitsuiYuki KuboNorihiro TakamuneDaichi KitamuraHiroshi SaruwatariYu TakahashiKazunobu KondoPublished in: Signal Process. (2021)
Keyphrases
- gaussian distribution
- nonnegative matrix factorization
- matrix factorization
- negative matrix factorization
- probabilistic latent semantic indexing
- maximum likelihood
- spectral clustering
- data representation
- machine learning
- nonnegative matrix
- learning algorithm
- face recognition
- objective function
- least squares
- gaussian mixture model