Direct density-ratio estimation with dimensionality reduction via least-squares hetero-distributional subspace search.
Masashi SugiyamaMakoto YamadaPaul von BünauTaiji SuzukiTakafumi KanamoriMotoaki KawanabePublished in: Neural Networks (2011)
Keyphrases
- dimensionality reduction
- least squares
- density ratio estimation
- low dimensional
- high dimensional data
- principal component analysis
- feature extraction
- high dimensional
- density ratio
- principal components
- unsupervised learning
- pattern recognition
- singular value decomposition
- data points
- semi supervised dimensionality reduction
- feature space
- data sets