Block-Active ADMM to Minimize NMF with Bregman Divergences.
Xinyao LiAkhilesh TyagiPublished in: Sensors (2023)
Keyphrases
- bregman divergences
- cost sensitive
- information theoretic
- maximum entropy
- nonnegative matrix factorization
- learning theory
- loss function
- exponential family
- mahalanobis distance
- kl divergence
- negative matrix factorization
- nearest neighbor
- theoretical guarantees
- data dependent
- boosting algorithms
- special case
- maximum likelihood
- clustering algorithm
- density estimation
- document clustering
- missing values
- mutual information
- multi class