Intrinsic dimension estimation by maximum likelihood in isotropic probabilistic PCA.
Charles BouveyronGilles CeleuxStéphane GirardPublished in: Pattern Recognit. Lett. (2011)
Keyphrases
- maximum likelihood
- intrinsic dimension
- intrinsic dimensionality
- dimensionality reduction
- parameter estimation
- principal component analysis
- maximum likelihood estimation
- dimension reduction
- low dimensional
- high dimensional
- feature extraction
- probabilistic model
- face recognition
- high dimensions
- principal components
- generative model
- expectation maximization
- em algorithm
- linear discriminant analysis
- density estimation
- high dimensional data
- principal components analysis
- bayesian networks
- covariance matrix
- singular value decomposition
- sparse representation
- maximum likelihood estimator
- face images
- noisy data
- diffusion maps