Sufficient Dimension Reduction via Direct Estimation of the Gradients of Logarithmic Conditional Densities.
Hiroaki SasakiVoot TangkarattGang NiuMasashi SugiyamaPublished in: Neural Comput. (2018)
Keyphrases
- dimension reduction
- principal component analysis
- feature extraction
- high dimensional
- high dimensional data
- feature selection
- low dimensional
- high dimensional problems
- singular value decomposition
- data mining and machine learning
- linear discriminant analysis
- variable selection
- manifold learning
- feature space
- dimensionality reduction
- probability density
- random projections
- cluster analysis
- high dimensionality
- discriminative information
- partial least squares
- dimension reduction methods
- sparse metric learning
- computer vision
- unsupervised learning
- data analysis
- high dimensional data analysis
- neural network
- density estimation
- nearest neighbor
- feature vectors
- pattern recognition
- intrinsic dimensionality
- feature subspace
- training data