Direct Estimation of the Derivative of Quadratic Mutual Information with Application in Supervised Dimension Reduction.
Voot TangkarattHiroaki SasakiMasashi SugiyamaPublished in: Neural Comput. (2017)
Keyphrases
- dimension reduction
- mutual information
- feature selection
- high dimensional
- principal component analysis
- feature extraction
- singular value decomposition
- high dimensional problems
- manifold learning
- cluster analysis
- data sets
- low dimensional
- unsupervised learning
- least squares
- image registration
- linear discriminant analysis
- partial least squares
- data mining and machine learning
- intrinsic dimensionality
- intrinsic dimension
- supervised learning
- high dimensionality
- pairwise
- reinforcement learning
- discriminative information
- similarity measure
- dimension reduction methods
- high dimensional data analysis