Condition for perfect dimensionality recovery by variational Bayesian PCA.
Shinichi NakajimaRyota TomiokaMasashi SugiyamaS. Derin BabacanPublished in: J. Mach. Learn. Res. (2015)
Keyphrases
- variational bayesian
- dimensionality reduction
- principal component analysis
- high dimensional
- factor analysis
- feature space
- matrix factorization
- independent component analysis
- exponential family
- expectation maximization
- dirichlet process
- incomplete data
- low dimensional
- density function
- feature extraction
- variational inference
- discriminant analysis
- em algorithm
- bayesian methods
- maximum likelihood
- mixture model
- high dimensional data
- feature selection
- linear discriminant analysis
- multi task learning
- density estimation
- data points
- face recognition